Subject: Internet Access Provider FAQ The Inet-Access Frequently Asked Questions List By David H Dennis (david@freelink.net) #html
The Internet Provider FAQ is kindly sponsored this month by

And

And

Check out a new book: How to Build an Internet Service Company!

## The FAQ was last updated on 22 September 1998 > Important Note If you're downloading a text version of this document from someplace or other, please note that you can always get the latest copy from http://www.amazing.com/ . Just follow your nose, and feel free to check out our other amiably eccentric web creations and resources. If you're just thinking about being an ISP, and you don't really know if this is the business for you, check out my advice for new ISPs at http://www.amazing.com/internet/advice.html . I now work for Freelink Communications, an ISP that now offers free advertising-based Internet access. You can visit us at http://www.freelink.net . If you need a web site designed, please visit my web design page at http://www.amazing.com/new . If you want to roll your own, you may also be interested in my web design FAQ at http://www.amazing.com/web-faq . Before I got my own connectivity, Avi Freedman of Net Access of Philadelphia was very generous in bailing me out when I was short on connectivity. I'd like to take this opportunity to thank him for standing by me. Check out what he has to offer at http://www.netaxs.com/ . > Introduction Welcome to the Internet Access Provider FAQ, 9,575 lines (and counting!) of information on how to become an Internet service provider. This is an amazingly broad subject, covering everything from politics to sexual mores to the care and maintenance of a large Unix site. We'll meet wanton women and straightlaced crusaders, lovers of Macs and of Suns, and even the occasional Windows NT user peering nervously in our direction. Yes, this is where we meet the traffic cops and maintenance people of the Information Superhighway, as they begin to make their first struggling appearances. They all have their idea of what the net is to be; about all they have in common is that they know it will be different from the way it is today. You can have many reasons for wanting to check out this FAQ. Perhaps you wanted to be an Internet provider from the day you were born, or maybe you're just curious to find out what it may be like. In any case, hello, and welcome to our world! This FAQ's for you, and I give a warm welcome to any comments and additions you may have. Write me at david@amazing.com > Every day in every way, this industry is changing This deserves some emphasis. Some people have defined a "Web Year" as a month or less in "real time". And this FAQ was started in 1994, when the web was barely visible on the horizon! So there's a lot here that I'm including just for fun and sentimental value. I hope nobody minds. It is, after all, my FAQ. :-) So you'll hear a lot about shell accounts and BBSs. I'll tell you right here and now that this information is largely obsolete; now, everyone's using graphical SLIP/PPP accounts off Windows or Macintosh systems. I've left the material in because I hate to eliminate anything that might be of interest to someone, somewhere. But, despite what I tell you elsewhere (in sections written many years back), the BBS and shell account business is no longer a viable one. Shed a tear for it. I did, since in many ways I liked it better; there was more room for individual creativity and thought. I went over the FAQ in considerable detail on June 2, 1997. There are new sections everywhere - I've brushed up information on Suns, improved the SGI section a great deal, added some more information on NT. I've also finally revised my "totally cheap" perspective to include information on the new access servers and 56k "standards" that have taken the modem market by storm. > Please do drop me a line ... but do me a small favour An increasing number of Windows [tm] users have been writing me using Unix-unfriendly mailers that do not break lines at 80 columns; each paragraph shows up at my site as one long line. My Unix mailer cannot digest these messages without a great deal of additional trouble, so please make sure that you use a text editor that can break lines at 80 columns or less (75 is a good number to use as a general rule). The easiest way to do this is to send your messages using a Unix Shell account with any of the standard text editors available. I realize, however, that this advice may seem too old-fashioned for the more enlightened and progressive among you. If you don't have Shell access, or if you cannot bear the thought of using Unix [tm], either: - Press [enter] at the end of each 80-column line as you are typing in your message or - Verify that your program DOES create [return]s at the end of lines. Spry's AIR Mail program seems to be particulary bad in this regard; I recommend that, if you must use Windows [tm] for some reason, you use Eudora instead. Microsoft Exchange under Windows95 is an absolutely horrid program that automatically generates obnoxious "rich text format" versions of each message you send, unless you explicitly turn it off. If you must send me mail using Exchange, please make sure those are turned off. Because their quoting conventions make text appear in a different font, and not prefixed with special characters, it is very difficult to follow a quoted message using Exchange. It's also best to switch off MIME format for messages sent to me, since doing so make it easier for me to read them. In reality, of course, the best thing to do about Exchange is to discard it and use Eudora or any other available mail client. Trust me on this one. I also have a great deal of difficulty reading mail with attachments, whether MS-TNEF (ugh!) or HTML. So please do not use MIME attachments if you can possibly avoid doing so; they'll make it harder for me to read your message, and therefore less likely to reply. > I'm thinking of making the FAQ strictly in HTML For many moons, the FAQ was just available as a text document. I'd really like to make the FAQ look better, by adding HTML formatting and such to it, but I can't do that easily as long as it remains available in text format. Does anyone lack the ability to read HTML nowadays? Drop me a line and let me know if this change would affect you negatively. > Some personal experiences I could make the FAQ even longer than it is now by just including my personal experiences in it. But because I wanted to amuse as well as inform, these are really in the style of stories more than FAQs. Check them out, and let me know what you think! David Buys a Router [ http://www.amazing.com/internet/router.html ]. In this gripping episode, we see our intrepid hero David Dennis visit the repair depots of the Information Superhighway in search of a router. Does he ever find one? Stay tuned. [Updated 16 November 1995; someone asked me for the brand and model of the router, so I added it] David gets an Internet Connection [ http://www.amazing.com/internet/connection.html ] At long last, David gets a 56k Internet connection. Read all about it! [Updated 28 August 1995] > Highlights of the Current Edition Section 6.11, Macintoshes running special MacOS software, has been updated with more information on using Macs in your ISP. I've also added a few notes on problems with Windows NT RAS. [Updated 10 November 1997] Section 10.5, Accounting and Billing Practices, now contains a comprehensive list of URLs to billing packages, as well as the detailed commentary I've previously developed. There sure are a lot of them, but unfortunately, virtually no information is included on most. Note that I do not sponsor or endorse any of these packages; I am providing the information because of a very high level of interest. [Updated 12 September 1997] Many, many sections were heavily revised and edited on 2 July 1997. About time, eh? A new history of Net-99 has been added, for the history buffs among us. I've added a nascent section on personnel. For the first time, Access Servers are now discussed in depth. Section 4.2, How do things pencil out? Some reasonably hard numbers, has been updated to reflect current pricing for T1 connections - and beyond. [28 September 1996]. Section 6.3, What about competition in local phone service?, has been added with information about Metropolitan Fiber Systems (MFS). Should you open your new business in a glossy downtown high-rise building? It might be your most cost-effective solution! [28 September 1996]. A new FAQ has been born, on running a web presence provider. Since it's brand new, you may want to study it and then return to this one for some of the details. You can access it at http://cgi.amazing.com/web-faq/ Section 10.7, Making the Internet more User-Friendly has been changed to include information on the pros and cons of supplying Microsoft Internet Explorer instead of Netscape [Updated 7 & 15 August 1996]. Section 5.4, Microsoft and AT&T, the Terrible Two, has had a name change; I previously mentioned IBM as a major threat, but it seems to have virtually vanished from the Internet map. I also bring you up to date on the Microsoft Network challenge, which also appears to be fading. [5 May 1996]. Section 9.2, Which News Software should I run?, has been updated with information about DNEWS, an intriguing alternative to INN. [5 May 1996]. Section 4.2, How do things pencil out? Some reasonably hard numbers, has been updated with more information on International rates. [5 May 1996]. Section 6.5, PCs running BSDI Unix, has once again been updated. BSDI users have fought back and responded to some of the bad-mouthing I'd added to previous versions of the FAQ. The complaints appear to be isolated incidents and not a major trend. [20 February 1996]. Section 15.0, Wanton Women and Straightlaced Crusaders, has been updated with fresh information on the copyright issue. [30 January 1996]. Section 18.0, Glossary, has been given a much-needed facelift with a bunch of important updates. [30 January 1996]. Another update to section 7.5, Who are the main national providers, and how much?, with some good news for ISPs. Net Access is going national. In addition, ANS, a corporate-oriented ISP known for first-class service, has finally lowered its prices to realistic levels. [Last updated 20 June 1997] I've added Section 7.6, What happened to Net-99/AGIS?, with detailed information on the rise and fall of this now-beleagured company. This section includes information about the legendary Spam King, Sanford Wallace. [Created 20 June 1997] The Internet and Windows NT: A summary of opinions. Should you consider this operating system as your next network server, or should you stick with the tried and true SunOS or BSDI? A satisfied NT user saw my earlier remarks, and replies. Oh-oh! There is now more information from an NT user and opponent, who brings us up to speed on problems with the Netscape Commerce Server under NT. More on this debate in Section 6.11, What about Microsoft Windows NT? [Updated 22 January 1996] I've created section 10.10, How has Windows95 changed the SLIP/PPP Picture. This will tell you much of what you need to know about Windows95 and its PPP implementation. Much to my surprise, it's not bad. The rest of the product has its problems, but PPP under Windows95 is a quantum leap past the horrors of Trumpet Winsock and friends. [Updated 6 December 1995]. For a couple of different views on setting up virtual domains, check out my Internet Resources page at a http://www.amazing.com/internet/ . Section 6.7, PCs Running FreeBSD, has been added to suggest another interesting operating system for ISPs. It has an excellent pedigree; how does it compare to Linux or other options? This section has been newly updated with Free BSD ISP mailing list information [Updated 1 October 1995]. I've added a very brief description of Firewalls in section 8.4, Firewalls. [Updated 18 July 1995]. I have significantly improved my section on Web browsers and servers, starting with section 9.9, What about Running a World Wide Web Server? This includes information on running a multi-homed web server (aka http://www.bigco.com/ [Note: That's a generic example, not a real link]) and comments on the Netscape extensions. [Updated 15 July 1995] Should you bother starting a provider with a 56k line? Maybe there's more hope than I would have thought, at least if the price is right. Check out Section 4.4, New Information on the Viability of a 56k line. [Updated 10 June 1995] Section 6.16, Care and Feeding of Disk Drives, has been started with a summary of the discussion on inet-access of overheating disk drives and what to do about them. Read this section so you don't need to feed your Barracuda to the fish! [Updated 29 May 1995] Ever been curious about Silicon Graphics, the super-upscale maker of fine workstations for graphic artists and their friends? Our spies snuck a peek at their Top Secret price list, and found that they're not as expensive as you might think. Check out Section 6.4, Silicon Graphics (SGI) Workstations. Any discussion of what's new with the Internet has to begin with Microsoft's announcement of the Microsoft Network, the new kid in the online service block. With their announcement of $ 4.95 a month access, they have thrown down the competitive gauntlet in a spectacular way. Check out our news analysis in Section 5.4, Microsoft and IBM, the Terrible Two. To Netra or not to Netra: That is the question! Section 6.17, All About the Sun Netra, tells you more than you ever wanted to know about this vital issue. Don't be deceived! Our FAQ maintainer decides to take the plunge and buy a Sun clone system. Unfortunately, thanks to a stubborn connectivity provider who just won't let loose an extra IP number, he hasn't managed to connect it to the net. He loves his Sun, especially with the 21" NEC "Professional Series" monitor he bought for it. But it has yet to be thrown into battle. Stay tuned. Our FAQ maintainer becomes horribly snobbish and spends more money than he should. Section 6.20, What sort of monitor should I get with my Sun? tells you what he did and why. > For Further Information, check out my WWW Site For further information on issues ranging from setting up Web Sites to using Linux as your Internet provider, check out my Web site, http://www.amazing.com/ . The latest edition of the FAQ can always be found here. The FAQ is updated on a somewhat occasional basis, about once every month or two. Visit the site and read in this page to find the last time it was updated. If you've been given a copy of the text version of this FAQ, you can see the latest edition in HTML by pointing your WWW browser to http://www.amazing.com/internet/faq.html . If you want to make a printout of the FAQ, a text version is available at http://www.amazing.com/internet/faq.txt . I would like to suggest that you visit the main site at least once, because the FAQ is only one of many interesting offerings we have. > The Obligatory Commercial David's Amazing Internet Creations is my new full-service Web advertising agency. Those who are long-time readers of this FAQ may remember that I ran a small ISP; since ISPs became more of a commodity business, I've shelved those plans and switched to the world of the Web in a big way. Since 1994, I had a small ISP that ran unique BBS software that I wrote; thanks to the death of the main machine running it, a Linux box, and the almost complete lack of user interest, it has been shut down. It was the wrong idea at the right time, alas. For details, please visit my commercial URL at http://www.amazing.com/new/ . > Table of Contents + Major Contributers to the FAQ I would like to thank the following people, whose efforts were particulary informative, helpful or encouraging: My thanks to Avi Freedman (freedman@netaxs.com) for sponsoring the FAQ once again and providing some much-needed resources. Karl Denninger (Karl@MCS.COM) has been kind enough to share full information about his equipment configuration with the list, as well as contributing intriguing observations on the politics of the net. I can't afford to duplicate his equipment, {sob}, and I certainly don't know how to dig into the swamps of Internet politics like he does, but reading about both is both interesting and useful. Draper Kauffman was kind enough to start an excellent discussion of the pitfalls and problems of operating a small IP provider. His discussion forms the core of my section on business problems, which I'm sure will be of great interest to many of you. Draper is the President of Illuminati Online, an Austin, TX provider. Alicia Salomon was kind enough to contribute information on many different topics, including organizational and administrative issues I have yet to become familiar with. She also gave me considerable moral support and encouragement when this project was just a gleam in its creator's eye. Alicia is an independant consultant working on Internet-related issues. In concluding one of her notes, she reminds us, "Also, something really important I forgot was you should include a pointer to alt.caffine." Consider it done, Alicia. :-) Bryant Durrell works as Postmaster for Netcom, and has generously offered to write the section on security for us. I particulary appreciate his input, considering the treatment I've given Netcom in various sections of this FAQ. Later in the document, I give him a manful apology. My information on connecting to the Internet via leased lines came mainly through the i.net info files (ftp to ftp.i.net:/pub/internetworks). My thanks to Aaron Nabil of i.net for putting together the clearest explanation I've seen of what equipment and services are needed to start up a full Internet connection. Several books I've read have not been as clear as his simple description of the requirements, as shown in his info-prices file. Most of the material in the BBS sections was kindly contributed by Mr Eric S. Raymond . Mr. Raymond is the programmer and designer of Chester County InterLink (CCIL), a free public access site in Chester County, Pennsylvania, USA. He is also the updater and maintainer of the legendary jargon file, now published as 'The New Hackers' Dictionary, Second Edition', a book I bought and read with considerable enthusiasm. (The book brings back all sorts of nostalgic memories of the good old days I spent hacking as a "Tourist" at MIT with the much-beloved and now lamented Incompatible Timesharing System (ITS).) This FAQ is copyright (C) 1994/1995/1996 by David H Dennis. Unlimited non- commercial redistribution is allowed so long as the text and copyright notice is retained in its entirety and not modified in any way. I will soon be splitting this document into multiple parts. You may distribute individual parts without distributing the whole, as long as the FTP site for the document is mentioned or some other way of getting the entire document is given. Commercial use is prohibited without written permission from the author. Release History and Changes First Release: Sunday 14 August 1994 13:20:26 FAQ written. Circa 527 lines. Second release: Sunday, 21 August 1994 12:52:10 Added distilled wisdom on low-capital IP startups; Added appendix on use of BSDI Unix as a router; added statement on Net-99 courtesy of Karl Denninger. FAQ balloons to over 2,000 lines for the first time. Third release: Minor changes, unrecorded by history Fourth Release: 17 November 1994 Added contributions on BSDI Unix by Tony Sanders and other suggestions from him, particulary much information on how to use a BSDI system as a router, and other generalized BSDI hardware recommendations. Added information on credit card acceptance. Rewrote the entire CIX section in view of recent changes to its status. A little more on 'Guerilla Financing'. A section on competition from the government, cable TV and telephone companies was added. Many detail updates were also made that I don't remember. Alas. :-( FAQ hits 4,119 lines, 197,862 bytes. Fifth Release: January 26, 1995 The hot new trend of marketing through the Internet, a new way to use your site to host stimulating and creative efforts, is now explored in a new section, thanks to the fine folks in the inet-marketing mailing list. Information about Sun and Sun clone hardware, including sidebars on the new Sun Netra server and an Internet "starter kit" is now included. Added contributions on Linux, security and system administration from Bryant Durrell , as well as an apology to Netcom lovers everywhere. Added the first version of my script to kill off runaway processes. Also added additional information for Macintosh and PC BBS lovers. An inspirational paen to our friends at Microsoft and Windows NT rounds up this month's exciting changes, and I chide people for trying to run Internet providers without learning Unix. FAQ once again jumps in size, this time to over 5,000 lines for the first time. Sixth Release: February 11, 1995 This release marked a major reorganization of the information presented, and an effort was started to convert it to HTML without completely messing up the text version. This FAQ was prepared using various computers and software, including a Tatung 85mhz SPARC 5 clone, a Sun 3/60, an IBM ThinkPad 750C, Sager and AST laptop computers and 486 DX2/66 computers running both Linux and OS/2. Both GNU Emacs (Unix) and Epsilon (OS2-DOS) text editors were used. The author most heartily endorses writing on the Sun/Sparc (when at home) and the ThinkPad (anywhere else); both are splendid machines for writing, perhaps the best available. Note added 24 June 1995 Gee, I haven't updated this section in a while, have I? I've more or less continuously revised the FAQ since I put it on the Web, but most of the changes have been quick additions which don't lend themselves well to a summary of highlights. In late January 1996, my 56k pipe started to fill, and the system is now running on a T3-connected site courtesy of Avi Freedman of Netaxs. Many thanks to him for his support! + Introduction +.# What is an Internet Service Provider (ISP)? An Internet Service Provider (ISP) is a company that connects members of the general public to the Internet. It can be distinguished from an Information Service such as CompuServe or America Online by its emphasis on Internet tools such as USENET News, Gopher, WWW, etc. Traditional bulletin board systems (BBSs) normally don't have direct access to the Internet and can offer only limited USENET news and mail, with no other Internet services. I consider any online system a provider that has a direct Internet connection and provides access to it in one or more of the following ways: * SLIP, CSLIP or PPP. These protocols let you connect your users directly to the Internet, using standard Internet protocols. Most importantly, this lets you use Netscape Navigator and other graphically oriented web browsers. Almost everyone currently using or running an Internet provider now uses this approach; the other two access methods are largely obsolete. * Unix Shell Account - Users are given the infamous "%" prompt or some variation thereof. Sometimes a simple menu is also provided, often cobbled up through various Unix utility programs. However, the basis of the system is Unix, and normally the user cannot take full advantage of the services offered without knowing at least a few Unix commands. Most people nowadays find Unix accounts to be horribly confusing. They are also significant security risks; hackers can do a lot with that % prompt that they could not do from their own machines. Shell accounts can also download information at the full speed of your connection (as opposed to your modem's). Finally, shell accounts take quite a bit more in the way of system resources because you have to offer them disk space and CPU time on your own computers. Because of this, most providers are rapidly moving away from shell accounts. Almost no providers even offer them anymore. * Provide customers with a custom BBS with specialized Internet features (newsreaders, etc). Many people have tried to put together some form of Internet access under commercial or shareware DOS or Windows BBS software. Most that I've seen have not been notably successful at this task. In particular, the lack of high-quality newsreaders for USENET makes the systems incredibly confusing to use. The advent of off-line mail readers may help this at least somewhat; the majority of postings through BBSs come from them. Unfortunately, most off-line readers tend to mangle headers badly and are a major source of annoyance (and occasional amusement) among USENET readers. The popular QWK format is a particulary bad victim of this practice; it uppercases all subject lines and limits them to 20 characters, both considered extraordinary poor form by USENET readers. The Whaffle DOS and Unix-based BBS has created a different off-line reading format which I believe may work better for USENET applications; unfortunately, I have not yet seen it in operation. Most Whaffle operators, however, find that users consider their systems confusing and difficult to navigate. Note that off-line reader programs, even those that work well with USENET, cannot help when it comes to using WWW or Gopher, which require real-time access to sites. TBBS, a DOS-based BBS program, has a system called the IPAD, which is apparently a 486/66 running some specialized software and including an internal router interface. It's quite expensive, but should give TBBS sysops a leg up when it comes to managing Internet connections. As far as I can tell, however, this doesn't help much when it comes to the messaging system, which is still ghastly. It's also not frightfully cost effective, as far as I can tell. Curiously enough, eSOFT has now discontinued TBBS and has based its entire future on the IPAD. I did some work under TBBS and found it an amazingly solid product; I'm sorry to see it go. Major BBS presently has an Internet module with severe limitations; for example, mail and news processing must still be done via UUCP (!). Based on the track records of the respective packages, I would assume that TBBS' package will be the better one. However, I think the best packages will probably be developed for Internet users by Internet system instead of existing commercial BBSs. Incidentally, we have the usual dispute over what names to give ourselves. Sean Shapira and others would like to call us Internet Access Providers (IAPs): "Personally, I call them Internet Access Providers. They are only one class of Internet Service Provider. Others focus on providing services to the already-connected Internet community. GNN and EIT are examples of this." Since the name of the mailing list is inet-access, he may have a point here. Why don't I change the name of the FAQ, then? Because I'd just finished changing all my ISVs (Internet Service Vendors) here to ISPs! :-) Anyone else feel strongly on this issue? If I get enough response on this issue, I'm willing to crank up M-X replace string again. :-) +.# Why might I want to become an ISP? Growth, money and the glamour of it all. The sleepless nights, the 18-hour days, the opportunity to exercise your mind and get creative in the provision of imaginative services for your users. The challenges removed in the move towards SLIP/PPP as a universal connection mode are largely restored by the creativity necessary to sell and create custom WWW sites, where considerably more money per sale is available. Despite all the publicity that's come up over the Internet, there are many rural areas that still don't have any form of connectivity. Many people now starting ISPs are doing so for idealistic reasons: either to provide good access in areas where most providers offer poor quality service, or to provide the only access in an area not yet served with a connection. +.# Why might I not want to become an ISP? Probably the most important reason is that competition is sharpening. Well-funded people such as Microsoft have discovered this business, and they've been highly aggressive and successful in the past. The small entreupeneurial company still has a chance to joust with the big boys due to the relatively level playing field, but the odds of success seem to be narrowing. The capital needed can be daunting to we start-up folks. True, it's nothing compared to setting up a CompuServe or America Online, but for the individual, it's a lot of money. For the group seeking venture capital, it's a lot of pain and paperwork. And then, once your dream starts being real, there are always the sleepless nights and the infamous battles with Sendmail and InterNetNews. The sinking feeling you might get when you're staring at a $ 2,200 router as part of a $ 3,000 a month connection, and realize that you don't quite understand the beast. The growth of SLIP/PPP accounts, which are very much a commodity business, much like the phone or cable TV, may change the creativity element beyond recognition for many providers. Still, the so far incredible growth rates should help make up for this. In an earlier version of this document, I mentioned the possible filtering of routes by the Commercial Internet Exchange (CIX) as a possible threat to your long-term viability as a provider. It's way past November 15th 1994 now (I'm writing this on February 6, 1995), and so far it appears that nobody's been much affected by the new changes. Finally, if your only reason to start a provider is to make money, you probably shouldn't do it. You'll be beaten out by those of us who love the net and who are willing to work utterly ridiculous hours to make your system a success. If you don't love staring at the screen for many hours, or if you value an offline social life, the net's not for you. +.# A Special Sidebar: Do I REALLY need to know Unix? I have received mail from a dismaying number of people who admit that they don't know Unix, and they don't want to learn Unix. They typically suggest that they use Windows NT or even Microsoft Windows to start a provider. To begin with, if you don't want to learn more about computers than you ever thought existed, you don't want to become an Internet provider. You'll find yourself staying up late at night reading thick computer manuals, instead of going out to that wild new night club you heard about. You may be reading this because you think being a provider is a sure-fire business idea; in reality, there are no sure-fire business ideas. You may think you can make a billion dollars with little work; lamentably, this is not the case, in this business or anywhere else. Still want to start a provider without learning Unix? Note, then, that the entire provider business, and the whole network of support you will be joining (including the Inet-Access mailing list) has standardized on Unix. Even trying to use another operating system for all but the most specialized purposes is well acknowledged in the ISP business as a colourful eccentricity and most likely a big waste of time and money. You might think that not using Unix would be like using Ami Pro instead of Microsoft Word, or using OS/2 instead of DOS and Windows. This is not a frightfully good analogy, since both OS/2 and Ami Pro are fine, fully functional programs. Windows NT does not have all the functions needed to work as an Internet provider; Microsoft Windows will crash the first time you tried to expire news. Unix is proven, easy to get help on, and has a far more comprehensive network of programs and utilities available. If you're not a real expert, you don't want to use a non-Unix operating system. If you don't know at least something about Unix, you'll have a hard time being successful as a provider. My recommendation is to become familiar with Unix through use of a shell account from a provider. You may find it easier than you think, in which case you're likely to enjoy running a Unix system. Or you'll think it's harder; in that case, you probably shouldn't be in this business at all. More details of the pitfalls of trying to run a non-Unix Internet site are given further on, but that should give you enough to think twice. + What about business organization and raising money? +.# What sorts of business organizations should I use? For the budding ISP, the sole proprietorship is certainly the simplest and most tempting organizational scheme. No tiresome forms to fill out, no lawyers needed, and no thought wasted on the issue. Many people, however, note the virtues of a Subchapter S corporation. This neat little format has the limited liability protection of a corporation, combined with the more favourable tax treatment of a sole proprietorship - you can still deduct losses directly off your returns. Unfortunately, the liability shield, while it looks great on paper, may not be quite so wonderful in practice. In the case of corporations owned by individuals, the courts typically "pierce the corporate veil", considering the individual and the corporation identical in case of liability. So if your solo venture goes bust, you may still wind up in trouble despite all your fancy papers. Thus, to make Subchapter S sensible, it's vital that you have at least one partner. It's worth noting that any corporation, including Subchapter S, requires a large commitment to maintaining curious rituals, such as annual meetings and elections of directors. If you forget even one small detail in these curious events, you may wind up losing your corporate status. Beware! Michael Krause summarizes the case for the S Corporation: * Tax benefits: "Give yourself no salary, take out loans from business, etc. Works out perfect, a nice little loophole in our tax system." [ Warren Henke [henke_w@cc.dixie.edu] comments as follows: "The IRS has become more strict on this point, and doing what you suggested has a very high chance of triggering an audit. The individual should be paid a fair salary, subject to SE tax. The remaining profits, however, can be creatively distributed. If you want rulings, regs, or other info, contact my firm: Savage & Esplin, LC, 801/673-6195; ask for Warren"]. * Personal Asset Protection: You can take out a business loan and leave personal assets protected - unless they are needed to guarentee the loan. Unfortunately, most banks are too smart for this little trick, unless your corporation has substantial assets and an existance distinct from your own. * decreased liability. Note that corporations cannot commit crimes; if your corporation is accused of criminal behaviour, you as owner are responsible. Like virtually all legal advice given on the net, this information comes from non-lawyers. You should consult your own legal experts before making any final decisions. +.# How much does it cost to set up an ISP? This is one of the most commonly asked questions. Unfortunately, it varies so enormously we can't give you any hard figures. If you're talking about a company using professional equipment, such as Suns and SGIs, access servers, and so on, figure on at least $100,000 worth of equipment and about $ 75,000 for the first year of technical charges (setup fees, T1 access, etc). In addition, figure on the cost of office space and people. You'll need at least one full-time technical guru if you're not one yourself. If you want to use Linux systems and access servers, figure on a minimum setup of about $ 35,000 for the equipment and - again - about the same $ 50-75k for the technical charges for a T1 and a healthy dial-up modem pool. None of this is cheap, and generally cheaper equipment won't save you nearly as much as you might anticipate. Unfortunate, but true. +.# What resources can I use to get capital for my ISP? Obviously, the easiest way to find money is to use your own. No tiresome investors to deal with, no people breathing down your neck when profits don't come quite as fast as one might like, and no delays waiting for people to make up their minds. So, winning the lottery should be the ideal way to finance your new business. Unfortunately, most of us will admit that this is rather a long shot. One solution to this problem is to have a full-time job already and live quite frugally. Then the balance of your income can be used to finance your system. I've managed to do quite a bit with this method, but unless you're used to a frugal lifestyle or have a princely income, it's really no fun at all. Worse, there are definite limits to growth; I can afford to fund my 28.8kbps SLIP forever; I couldn't afford even a month's worth of T1 charges without outside help. Unfortunately, as we will find out a bit later in full detail, this kind of plan may not be enough to start a profitable provider. Parents, friends, relatives and people who have seen you in action are probably the best early investors. They know and trust you through previous dealings with them. Creating a preliminary business plan and walking through it with them will not only help persuade them; it may also yield good advice about what is needed to sell it to them or others. (My most likely source of financing is my day job boss). A recent book called 'Guerilla Financing' tells how you can locate "angels", individuals keen on investing small amounts of capital ($10-100k) in start-up ventures. Their motivation often comes as much from the glamour of being associated with some prestigeous enterprise as a desire to generate profits; this is, for example, the spirit in which Broadway shows are traditionally financed. The book has some fascinating detail on the precise procedures, and I will have to buy it next time I see it to give some ideas and better pointers. In sum, however, an Internet provider business might be a particulary good field to find an "angel", since they are often interested in the glamour of investing in high-tech ventures. There are other, more formal, sources of investment. Venture Capitalists. They take a sizable percentage of your business for the money, but they might be your best chance. This is especially true considering the "trendiness" of this subject. (Someone who has actually dealt with a venture capitalist might be able to help me out a bit with this section. :-) ) Note that venture capitalists are almost always interested in eventually taking your company public and reaping the sizable returns that can be obtained therefrom. If you don't want to eventually wind up with the hassles of running a public company, you may not want to go this route. Keep in mind, though, that it could make you a wealthy person if everything goes right. Netcom went public a few months ago, and Bob Reiger, its founder, is sitting on a mammoth paper profit right now. (Note that this is true even though Netcom has been a perenial money loser). Banks. Not really an option until you're a going concern. The US Small Business Administration (SBA). According to Joseph Lamar Greer , it is nearly impossible to get a SBA loan unless you can qualify as a minority. Many people have gotten loans by taking on a qualified minority partner or partners. Tom Berdan is an Executive VP of a small bank in Tampa, FL. He has been a SBA lender for the past years, and wants to correct some common misconceptions about the SBA. He writes us as follows: "Your section on 'What Resources Can I Get For My ISP" quote a Joseph Lamar Greer as saying it is nearly impossible to get a SBA loan unless you're a minority. This is the single largest misconception of the SBA loan programs. In fact, the largest program is the SBA 7a program, which provides loans to SBA businesses. No minority qualification is required, and if the business owners are miniorities, it does NOT factor into the decision process. SBA loans last year totalled over $ 7 BILLION, with projections to approach $ 9 BILLION this year. "The basic requirements for a SBA loan are that the owners must have some capital, management expeirence, demonstrate that there is sufficient cash flow to repay the loan, and collateral. SBA loans probably are difficult to find for $ 10,000 to purchase the equipment for asmall 28.8 ISP provider, but oculd definitely be available to an ISP after a successful track record as a small ISP looking to expand. It really depends on the lender - which is where one starts the process. The SBA has recently started a low doc loan program. This program is designed for businesses needing less than $ 100,000 in loans, and the amount of paperwork submitted to the SBA is greatly reduced. However, the same 'basic requirements' above still apply, and NO minority participation is required. My advice to anyone interested is to call their local SBA office and ask them to provide a list of lenders in the area. The process starts there, with the lenders, and not with the SBA. Also, the SBA is available on the Internet. If you're interested, drop me a note and I can provide you with the HTTP page address." [It sounds to me like this is a great deal like a regular bank loan, in that your financial viability needs to be assured before you can successfully pursue this option. As Tom says, the SBA gives loans, not grants or direct investments; as a result, there needs to be a very clear and straightforward path to repayment. I don't think this is something most beginning ISPs have; however, it would certainly be useful for expansion, as he says]. Lottery Funds and other Economic Development Grants: In some states, grants are available from the state (often funded through lottery proceeds) for business development. Note that this is in the form of grants; you don't need to pay this money back. Department of Commerce: The US Government is investing $ 1.2 billion a year to promote "development and deployment" of the "National Information Infrastructure (NII)". Whether this means anything that can help us is open to question, but you can check out their FAQ on this subject through their BBS (202) 482-1199, or their WWW page: http://ntiaunix1.ntia.doc.gov:70/0/faq/niiques.asc You would presumably have to be a non-profit or school to get these grants. + What sorts of returns can I expect making as an ISP? +.# Introduction: The world is changing. For quite a while, you could start an ISP on $ 10k worth of equipment and a smile. Karl Denninger's MCS-NET (mcs.com) did that and became a sizable presence in Chicago very quickly. Avi Freedman's Net Access started on a 14.4kbps line, an ancient SPARC and a couple of modems. Congratulate them. It's not so easy anymore, especially in areas with present competition. You probably have your best shot if existing ISPs have dreadful reputations. In fact, I decided to start my provider after noticing Netcom's bad service; Netcom is the main provider I know of for Southern California. (When I started my provider service in Van Nuys, California, in the San Fernando Valley, I was the Valley's first "native son" provider). When I started this FAQ in 1994, Netcom was pretty much the only game in town, and despite lots of customer gripes, it was growing at tremendous rates. At the time, Byrant Durrell of Netcom, who has since left them, said that I was being unfair in regard to Netcom's quality of service in the previous paragraph. (Now, he has long since left Netcom). In November of 1996, while I'm writing this update, Netcom now has as poor a reputation as ever. More national providers have joined Netcom in chasing the elusive goal of national service. Unfortunately for them, most of them have the same reputation Netcom does. Apparently, it's expensive to get customers on a national basis, and it's also costly and inefficient to run national networks. What this means is that, in markets not yet saturated by local providers, the provider opportunity is still there and most likely worth pursuing. My thanks to Draper Kauffman for providing the initial inspiration for this section. +.# How do things pencil out? Some reasonably hard numbers At long last, I've done a healthy amount of updating to this section. One of the most controversial aspects of Internet provider lore is how much load you can put on your system and its Internet connection before things become intolerable to your customers. To run a high-quality, conservative service, the consensus seems to be the following: 28.8K SLIP: You can run three phone lines and get reasonable, but not great, service for all of them. This is what I was doing in 1994, and it did work well, although the lag in telnet connections (caused by high latancy on the phone lines) is bothersome to many. CSLIP (compressed SLIP) seems to produce a dramatic improvement over SLIP. 56k: You can run up to eight phone lines and get satisfactory service. However, see our new section "What about a 56k line?" for some details and some evidence from 56k defenders. For a long time, I had a line, and I must inject a word of caution about these figures. I presently have six phone lines and a very lightly used BBS/ISP (time has really passed it by, due to its text-only interface). For about six months, I had my voluminous web pages on it. It took me under three months for my web pages to completely saturate my 56k connection. So if you're planning to offer any kind of web page service at all, you need a T1 line or better. As a result, for a long time I had my web pages on another machine, thanks to the generosity of Avi Freedman of Net Access. He has a T3 line, so bandwidth is no longer an issue. But you can see that offering web pages on a 56k line is pure folly if you expect them to enjoy any level of popularity. ISDN: Some people have considered starting an ISP using an ISDN connection. Depending on your location, this could be a decent idea or an extremely stupid one. It turns out that most telephone companies that offer ISDN are charging by the minute for the service. At $ 0.01 per minute, that's $ 432 a month, or about the same as the cost of a full T1 frame relay connection. So if that is the case in your area, you definitely should not go this route. There is a possible way around this called (in Pacbell-land, anyway) Centrex ISDN. It's available only if you're in the same phone prefix as your upstream provider. That is, if your number begins with 818-997, and your upstream provider's number begins with 818-997, you could be lin luck. If you qualify and can get it set up (which is apparently one of those painfully ardurous processes), it should work for you. I have been told that dealing with the phone company on this is even worse than dealing with them on a regular ISDN circuit, which is horrorific enough. If you have flat-rate ISDN, or if you can take advantage of Centrex ISDN, it's a good, cost-effective way to get started. But remember that ISDN equipment won't go above 128k, so when you grow, you'll have to buy all new equipment. An ISDN router, which you'll need as an ISP, is about $ 1,000. T1: Now we're motoring! A T1 line can support up to 27 times more lines than a 56k - thus, around 200 phone lines. But that doesn't tell the whole story; a recent discussion between users on Inet Access with more than a T1 reveals that a T1 should be able to support the actual throughput of around 300 users. Here are some rough rate calculations, together with the provider that supplied them. This pricing information was updated on 28 September 1996. ------------- --------- ------------ --------- ---------- ------------- Provider Service Monthly Cost Line Cost Total Cost Cost/T1 Cinenet 28.8 SLIP US$ 125.00 20.00 145.00 Absurd Various 56k US$ 350.00 100.00 450.00 Net Access T1 US$ 600.00 500.00(1) 1,100.00 1,100.00 ISI T1 US$ 918.75 500.00(1) 1,418.75 1,418.75 ISI 10mbps US$ 2,726.25 (2) 2,726.25 419.88 UUNET T1 US$ 3,000.00(3) 500.00 3,500.00 3,500.00 UUNET 10mbps US$ 6,000.00 2,750.00(4) 8,750.00 1,347.61 UUNET SMDS 34mb US$15,000.00 Unknown Unknown Quite Low UUNET T3 US$49,000.00(5) Lots Massive Unknown ------------- --------- ------------ --------- ---------- ------------- (1) Actual prices for T1 connections using your local phone company will vary dramatically. Pacific Bell and GTE prices for Southern California vary from about $ 400/month to $ 1,000/month to pipe a T1 to your desired location. Net Access is a east coast-based provider based in Philadelphia, PA. It is the connectivity sponsor for this FAQ. From all the comments I've heard on the Inet-Access mailing list, it is one of the few wholesale access companies with an absolutely spotless reputation. Do business with them if you can; you'll be glad you did. http://www.netaxs.com . (2) Requires the rental of space in the same building for a direct Ethernet connection. It's also possible to use MFS On-Net service to connect to other MFS-connected buildings at the $ 2,750 rate below. You can also pay the monthly charge at a flat-rate $ 30,000 a year, with a further 5% discount if you pay within 15 days of invoice. (3) Substantial discounts from these published rates are rumoured. The list price of installation is $ 8,000, but substantially lower rates (circa $ 1,000) have been reported. The actual price is dependant on usage; I've shown you the highest tier, so with some luck, actual pricing will start low and build with revenues. See http://www.uu.net/pricing.htm for full details. Additional Discounts can also be had with a term commitment (one, two and three-years). (4) Requires tenancy in a MFS On-Net building, usually large office buildings in major metropolitan areas. I previously included Net-99 in my price lists. Because Net-99 effectively no longer exists, it has been removed. AGIS, Net-99's successor company, has been mainly notable for horrifyingly poor service. Rest in Peace, Net-99. Cinenet is the provider I used for 28.8 SLIP at the time I wrote the first version of this FAQ. I currently have a 56k connection with via.net [ http://www.via.net ]. ISI is the provider I use for T1+ services. They are a university consortium based in Los Angeles. You can find them at http://www.isi.edu . UUNET provides what is probably the best service of the major providers. MCI is also looking good for service quality, but watch out for their rates and billing. They are offering a tiered rate system which is said to be quite expensive compared to their previous costs. Connection Phone Rate/L Cost/ Cost/ Gross Connection Cost/Line Lines Commercial Line User Income/L Total ---------- --------- ----- ------------ ----- ----- ------- -------- 28.8 SLIP 48.00 003 30 (bus) 78.00 7.80 100.00 300.00 56k 56.00 008 30 (bus) 86.00 8.60 100.00 800.00 T1 12.00 200 30 (bus) 42.00 4.20 150.00 30000.00 ------------- --------- ------------ --------- ---------- --------- For a long time, European and Asian rates were as much as ten times higher than US rates. According to Antti Summa [anti.summa@netropolis.be], this is no longer the case, with rates for major urban areas about the same as in the above table, and in rural areas around 2-2.5 times more than that. (If this is true, it would make European rural rates lower than American ones!). I'd be interested in hearing other non-US provider's experience with rates, so I can update this information better. Here are some sample European rates, kindly contributed by Hans Michalec : FYI the recent rates for the data lines in Austria (via EBONE). Rates are in ECU (approx. =1$)/year. So, a 64 kbps line does approx. 33.000 $ a year - LUCKY AMERICA! ----------------------------------------------------------------------------- speed/kbps: 9.6 19.2 64 128 192 256 512 1024 1536 2048 3072 4096 commercial cost/kEcu: 14 20 33 60 91 118 206 292 376 442 694 832 *) academic cost/kEcu: 8 11 15 27 41 53 92 122 153 174 275 320 ----------------------------------------------------------------------------- *) commercial members whith their own US line get a reduction of 3 kEcu per 64 kbps access. (end of quote) When I first wrote this document, SLIP/PPP accounts were an exotic rarity and most people used the net on a Unix shell account. Nowadays, most people are using SLIP/PPP accounts, so comments about the older shell/BBS are mainly obsolete. Because of the proliferation of providers, SLIP/PPP accounts have gone down dramatically in price, to about $ 16.95 unlimited access. However, the industry has moved into two tiers: Mediocre service for $ 16.95 or so, and first-class service for around $ 25-30. As a general rule, beginning ISPs should plan on offering first-class service and charging rates on the higher side. If there is a real proliferation of providers in your area, Internet access may no longer represent a first-class business opportunity unless you have some special "hook" with which to attract customers. At this point, I would strongly discourage you from trying the access provider business unless you can afford a T1 at minimum - unless you're planning on supplying access in a rural area with a minimal customer base. If you are doing that, however, be prepared for extremely low potential profits depending on your community's status. The cost per user assumes 10 users per line, the standard. Gross income per line assumes rates of $ 10/month for 56k or below, going up to 15/month for T1 service. Running the numbers shows extremely low gross profit for 56k or 28.8 SLIP connections using the standard measures. Of course the T1 user can also be a SLIP account, which commands about a $ 17/month or more fee. So you can see how much better a T1 is than a 56k; you can not only sell more accounts; you can sell more expensive ones. To make matters worse for the 56k connection, the standard of 10 users per phone line is effective for about 16 phone lines and up. For fewer lines, you may be talking about 7-8 users per phone line, because the concentrations of users tend to clump more for a small number of people. Louis Epstein [lepslog@j51.com] reminds us that in some areas, a fractional T1 can be a good investment. Where he is, for example, he can get a 384k connection for $ 1,600, while a full T1 would cost him $ 2,200 a month. I had talked to people about fractional T1 service before, and my impression had always been that it was not that good a deal compared to a full T1. This example doesn't exactly refute this; for only 50% more a month, you get four times the bandwidth. But if you're in a tight budget, and don't have use right now, it sounds like an option well worth checking out. In the case of his provider, they will upgrade to the higher speeds for just the price difference between the fractional and full T1. So when you're ready, it's easy to switch. This is a big black mark against the 56k, whose service needs to be completely changed and re-installed when you convert over to T1. 28.8K SLIP deserves mention as a viable option for the start-up provider, just to get its personnel familiar with the various pitfalls of running a system connected to the Internet. I've learned a great deal with my 28.8, and recommend it very highly to anyone considering a start in this business. As you can see by the table above, it's no moneymaker, but it's not a dramatic money loser, either. Anyone with a more or less decent daytime job could maintain a 28.8 SLIP provider for as long as their interest continued. The main problem with a 28.8 SLIP is its main advantage: You do not use a router or a CSU/DSU. It's good, because you don't need to cough up the cash for them. But you're not going to get to know one, either, and that can be a problem when you upgrade to a better connection. Another disadvantage is that you can't distribute the load between machines; SLIP connections are pretty well confined to a single unit. (In 1994, when I ran such a connection, I had this problem; expiration of news was taking a long time on my system, so I would have benefitted greatly from having a news server. Alas, I can't hook one up because I'm just connecting a single machine and not a network). The 56k option looks like an extraordinarily unhappy compromise. If we subtract basic expenses of $ 86/line from the gross income of $ 100/line, we get only $ 14/line in gross profit, or $ 112/month for an 8-line system. Obviously, nobody's going anywhere on a 56k connection unless they either charge a lot more than the going rate as I know it, or bend the rules to the breaking point. However, it certainly is a way to get to know your router and CSU/DSU without spending massive amounts on them. Finally, if you are lucky enough to be able to afford a T1 connection to the Internet, and have a successful marketing plan, the T1 option is very clearly an excellent viable business. If you could fill up all 200 lines a T1 is capable of serving at a rate of $ 15/month, you'd have 2,000 users and $ 30,000/month to dispose of. You'd probably need to hire a few employees at that level - it couldn't be done properly with just one person - but you'd be able to afford them. At this level, you could also resell 56k network connections to companies that needed their own net presence. Joe McDonald says that this is a surprisingly easy thing to do, and should be considered in any projections involving T1 lines. Tony Sanders suggests fractional T1 as a good way to 'ease in' to the real thing. "You pay full T1 line charges but the total cost is quite a bit less and it makes for a very easy upgrade path." This seems to depend a great deal on both the provider and the telephone company you have to deal with. When I was talking to ISI Network Associates and Pacific Bell, I found that ISI charged the same for Frac-T1 and T1, and Pacific Bell only charged about 20% less for the smallest increment of frac-T1 (256k). T1 people can also sell SLIP connections reliably, which are generally significantly more expensive (in the $ 20/month and up category). SLIP is generally a high-bandwidth eating operation, so normally you won't want to sell this form of connection on a 56k or lower line. However, note well: Joe McDonald has successfully sold and operated SLIP connections on a 64k Frac-T1; he says both he (when he calls in from home) and his customers are happy with the service. Craig Warner ; http://www.pcisys.net/~craigw sells an Internet Starter Kit for a bit over $ 20,000 (see the section on Sun hardware, below). He says that this should support 1000 users before upgrades; projected break-even point is around 700 users. "One thing to keep in mind is that business accounts can generate a great deal more over individual accounts - we paid about $ 200 a month - with only a dial-up UUCP connection until recently. We now have a 56k Frame Relay connection which costs about $ 400 per month in total fees." The aforementioned Joe McDonald charges $ 200/month for permanent 28.8k SLIP; he already has one such customer after only two months of operation. +.# Co-Location with another ISP If you are doing primarily web page provision, without user accounts, by far the cheapest way to get started in an ISP business is co-location of your server at another ISP's site. Ideally, this company should be a large, T3-connected site, although it's possible to get "too good to be true" deals if you do business with someone who only has a T1. Avi Freedman of Net Access, for example, charges $ 800 per month per machine, for T1 bandwidth or below off his T3 link. This is far cheaper than any other T1 solution, and can give you extremely high speed access, possibly better than if you got a T1 from a smaller company. Co-location with a small ISP can cost as little as $ 200-300/month, but you may find that the bandwidth they have is not adequate for your purposes, especially if they're co-locating a large number of machines off a single T1 line. If you're a small, rural ISP with only enough capital for a 56k or 128k line, a particularly intriguing idea would be to run your own direct access site off the 56k, but co-locate with a large, urban company for your web sites. That gives you the ability to sell high-speed web access, without the need to pay for expensive phone connections to your actual site. +.# What if I oversell my connection? Probably the most tempting option for the 28.8K or 56k provider who'd like to get some decent profits out of his system is to oversell the connection - that is, to exceed the recommendations listed in the previous section. The argument is seductive: Many people are cheap. They'd rather have an inexpensive connection than one that worked perfectly. I (the start-up provider) am just one person, and I can't provide a perfect system in any case; I just don't have the capital for a T1 or a 24-hour staff. Could I play the ISP game anyway, by just selling an overstressed connection for less money than other providers? The main problem with this game is that it's too easy to play. If you offer cheap service at cheap prices, there are bound to be people with more resources than you who can offer cheaper service at even cheaper rates. This is Karl Denninger of MCS.COM's comment: There will always be someone who can undercut you. Quality service, though, is very hard to come by in the Internet provider business. Many very large providers, such as Netcom, are highly vunerable to complaints about terrible service. If you can capture some of their customers by offering excellent service - even for more money - you probably have a very good shot at gaining market share, even over very large companies. So you may want to at least consider the high road, not the low. Craig Warner adds: "As a case study, our provider failed to upgrade his hardware to keep up with growth - they lost over 500 accounts in a month." Michael Krause is an excellent example of a provider who's managed to succeed despite the handicap of a slow link to the Internet. His system runs 8 phone lines on a 28.8K link. What makes this possible is that only a few services - most notably FTP and downloading images through WWW (which shell or BBS account people can't do anyway) require a major proportion of the data pipe. Mudders, Telnet and IRC users are low users of the system capacity. Usually he sees a maximum of two simutaneous FTP sessions, which can be handled by the system. He points out that, as long as his customers don't have experience with faster providers, the speed is not a tremendous issue with them. He suggests that the peak number of users on a 56k would probably be between 75 and 100. A couple of other people have mentioned 40 as a "reasonable" maximum number of simutaneous connections on a 56k with light FTP traffic. Note that heavy use of FTP and WWW may change this dramatically in time. The aforementioned Craig Warner suggests: "Intelligent local caching is another way to survive with lower bandwidth. A 3-5GB cache on a server could cut bandwidth requirements significantly." From personal experience, I know that users will stick around after just about any disaster, as long as access is free, and it's understood beforehand that the system is experimental. I knew that my system would be unreliable due to the new software I was writing, and my inexperience as a system administrator. As a result, I started by running it at no charge, with the understanding that lost mail, connection problems and such would be accepted as typical experimental system pains. So far, the overwhelming majority of my users are very loyal, but that may change when I start charging even a nominal amount of money. In short, giving service with problems at a low price may be a reasonable strategy, but unfortunately there's a big difference between "low" and "free" in most people's minds. A couple of my users have told me that they don't want to pay for the system as long as telnet connections are so slow, so even my strategy may not pay off. This may, however, be due in part to problems with runaway processes I've had on my system, which have decreased performance for all users. Since then, however, I've solved my runaway problems and gotten a CSLIP connection, which is much faster, and I seem to have silenced the doubters. News and mail reading and writing, of course, use virtually no bandwidth at all, and you could run a news/mail only system on a very slow SLIP connection. However, the value of this to the general public is questionable. Still, if this is the primary interest of your users, they'll probably be pleased with virtually any bandwidth level. Note, however, that even a 28.8k SLIP doesn't seem to be enough for a full newsfeed through INN, at least through my present provider. (Things have improved dramatically since I started using CSLIP with them, but how much I'm not sure at this time). Many people get Internet access through their work or school, but those institutions normally censor the content provided. For example, it's pretty tough to find alt.sex.bondage on a corporate machine, unless you're its manager. This opens up a surprisingly large market of people who have access already, but want to telnet in to check out the "forbidden" topics and sites out of their school or employer's wary eye. This is likely to grow with the recent decision at a major university (I think it was CMU) to halt access to sexual newsgroups. This might make a system with an excellent news connection and nothing else a viable site. It's also an excellent way of competing with the Freenets, which are cheap but heavily censored. Many providers, including MCS (Karl Denninger's provider), offer cheap telnet in accounts for that reason. Because the users are just using a small portion of your network connection and no phone lines, you can charge them a lot less for access and still make money. Unfortunately, the "forbidden" topics, including the alt.binaries.* newsgroups, have increased dramatically in size, making a strategy created around them probematical unless you have either a T1 line, Pagesat service or both. I have noticed that, as long as FTP works, the slow speed doesn't bother me much. What does bother me is delays in character echo when I type. This seems to indicate that, if your system hardware is fast enough, you may be able to deliver service that's perceptively better than Netcom's even with quite a sluggish link. My conclusion to all this is that many ISPs can get away with bending the rules for a while, at least until the competition gets a T1. Then, all that careful business planning and development may go to naught. As Tony Sanders summarizes the situation, "It's kind of like a race for the T1 connection :-)" +.# New Information on the Viability of a 56k line Several people have written recently that I have significantly understated the potential of a 56k line. Apparently many people are running between 10 and 20 simutaneous users off a 56k with considerable success and happy customers. The reason for the discrepency has to do with the usual use people put to the line. In a traditional Internet provider, a large percentage of users were engaging in FTP sessions; the minimum allowable bandwidth was based on a large percentage of users continuously FTPing stuff. Now, most people who were formerly FTPing are using the Web. Fortunately for providers, this means that most of the time they are just looking at documents, instead of sucking them up and departing. As a result, people stay on your system longer, but use less of your bit pipe. Christopher X Candreva writes: "I'm currently running off a 56kb: Full news feed, 20 dial-in lines. It's just getting tight. Large FTPs are slow, but PPP people are still limited by their 28.8 modems. The only people who complain are shell FTPers. "And I'm upgrading to T1 by the end of the month." Jacob Westfall has a similar experience. "I run a small ISP with SLIP/PPP users. We have 20 incoming lines and get 150M of news a day. Typical transfers for the 14.4 users is 1-2k/s. Most of them use WWW and speeds on those lines are pretty good. Most pages load in a couple of seconds, the longest page to load I have seen was 1 min. The only complaints about speed I have ever had were first time Internet users who have just finished using a BBS and complain that they aren't getting the same transfer rates. Some of my users were signed on with larger Internet providers who have full T1s and they say the speed with my company is the same as with their previous provider. The main issue is not overcrowding the line. 20 lines on this 56k is as far as we are going. Interesting note: Our provider has a T1 and says he is only using about 256k of it. What most people have to realize and this is sysadmins included is that IP is not constant traffic. Watch the lights on a modem transferring IP sometime and you get the drift. 80-90% of the time the lights are idle while customers are using the web. What you have to look for in a Frame Relay 56k which is what we have, is that the provider you get it from has a large enough link into the cloud. Our provider had 128k into the cloud and just doubled that to 256k. Our speeds have almost doubled in terms of NNTP transfers from them and overall user speeds have jumped about 20 percent." Note, however, that many of those people are in places such as Canada where T1 lines are prohibitively expensive, and 56k service costs almost as much as T1 here. Still, even in the US it would appear that you can get a 56k line to work as a provider, as long as your service offers a stripped down newsfeed. A full newsfeed of around 450 megabytes a day is definitely not going to fit well within a 56k. Before December 4th, a company called PageSat offered a USENET feed via satellite; the company's reputation was always a little flaky, but customers coped with it due to the sheer brilliance of the concept. Unfortunately Pagesat now appears to have gone out of business. I have more detailed information on how this service worked and its demise later on in this document. Note that if your newsfeed is slow, you may have the equivalent of a partial newsfeed without realizing it. INN will not use all of the data pipe for transferring news, at least not in my experience. As a result, my system, running off a 28.8k SLIP, probably isn't getting anywhere near a full feed, even of the groups (alt, rec and misc) I'm getting. This may make your 56k or lower connection appear to work better than it actually does. Despite this apparent good news, everyone I've seen is going to a T1 or trying to do so. Once you hit that 20 simutaneous user barrier, you're almost certainly going to need one, and need it yesterday. If you can afford a T1, get one by all means. But if you're on a shoestring, and have some way you can compete with other providers in your area, a 56k just might do the job. If nothing else, it'll tell you if there's demand for your service before you take the plunge and start spending the big money. +.# Draper Kauffman's 56k Connection Sudden Death Scenerio This message is both so scary and so dramatic that I can't resist keeping it in the FAQ, even though I've summarized many of the financial arguments above. The message is that if you have a 56k connection, you have at most a $ 500 gross profit, even with a $ 20/month rate. And $ 20/month is unsustainable in today's market. Here's what that $ 500 would cost you, if you could even get it at all: From Draper Kauffman : How hard do you have to work to get that? To begin with, if you are growing fast you've recruited a lot of new users recently. They'll have a million questions and requests. You're also doing all the accounting, putting out accurate monthly invoices for 100 erratic users, depositing 70-80 checks, and deciding what to do with the no pays and late pays. Meanwhile, you're trying keep your system up and your 8 bargain modems working, and deciding what to do when mail runs 2 days late, or your newsfeed stops coming, or whichever of the normal Net-crises hits you that day. All told, you probably put in 50-70 hours per week, maybe a lot more. You make less than a ditchdigger, and you aren't getting paid for the use of your equipment, so you haven't got any source of cash for upgrading or expanding your system. But you're still adding customers! And here comes the crunch that threatens almost every low-capital ISP these days: too many users asking too many questions, not enough bandwidth, people bitching about how the system slows to a crawl every time you get your newsfeed, programs crashing because the user disk is full, and so on. Too many problems are allowed to fester and turn into flames, and suddenly you just can't cope. There aren't enough hours in the day or dollars in the bank. Angry customers quit and bad-mouth the system and new people stop coming. If you keep your rates at a moderate level you won't have enough volume to cover your costs. If you drop your rates to bargain levels, you keep more customers, but you're losing money on every one, so that's no help. You're bleeding cash, and pretty soon some unforeseen expense will put you out of business. You probably won't even know it until your check to the telco bounces, since you haven't had time to do the bookkeeping for months. That's the nightmare. Even if you can raise more capital at that point and try again, you have to fight the bad rep of having run a shoddy operation. And bad word of mouth hurts you more on the Net than in almost any other business. What causes this scenario? Here are ten good reasons: 1. Having insufficient capital. 2. Underestimating the time and routine expenses involved in a startup. 3. Overestimating the owner's knowledge, abilities, and stamina--the heroic programmer complex. 4. Grossly underestimating the manhours needed to get the system up and keep it running smoothly under load. 5. Starting with inferior services: slow connection, inadequate disk space, skimpy software (gopher, lynx, etc.), slow or partial newsfeed, and/or running too many functions (news/mail/users/etc.) on one CPU and drive. 6. Charging too low a price (as a result of points 2 through 5). 7. Using cheap hardware and no backups. 8. Starting with a system with insufficient capacity to produce enough profit to finance continued growth. 9. Allowing growth to exceed the sustainable system capacity. 10. Sloppy and inadequate accounting--it's easy to get behind, and usually fatal. (I would add inadequate marketing to the list, except that that can easily be a blessing. One highly successful marketing effort could add 2-300 users in a week, completely overloading the system.) Here's Draper's summary of the present environment: Although there is market to market variation, today's ISP startup faces a significantly more difficult challenge than those that started in the last few years. Quality expectations are up and user fees are down. The result is that low budget/low volume/low quality/low cost systems can no longer expect to find a profitable initial niche in most markets. Without new capital or a steady stream of profits, they have no way to increase quality or capacity Increasingly, a new service in a competitive area needs to offer a full range of services, good user support, and a fast, reliable, and accessible system in order to charge a premium price. Success will require larger amounts of capital or inventive ways to overcome the numerous barriers to low-volume profitability. +.# So, what can we do about this? These are my personal suggestions as FAQ maintainer on what to do if you're a low-budget IP startup and somewhat intimidated by all this: (1) Create some innovative services. For example, you might want to host community-related sites, such as the police department and local charities. You could also spend a few hours a week net surfing and reading net oriented publications like Wired. Then, you can announce the "newsgroup of the week", "URL of the week" and "Telnet Site of the Week". It would also not be a bad idea to put out a monthly newsletter that contained that information and told people to be sure to come online for that and similar events. I think if you make your system a valuable resource to show people what they can do on the net, you'll build loyalty that will pay off when the crunch comes. And you don't have to do that just through direct personal communications; inexpensive media such as a simple informational menu and newsletters will do fine. (2) Try to find a backup source of capital. From what we've seen in previous sections, it's highly doubtful that 56k is a viable solution, especially if you need to share your returns with investors. As a result, it's pretty much T1 or nothing. Costs for this seem to vary dramatically depending on your market, as I've covered previously. (3) Don't quit your day job. A 28.8k system will work fine for you to test your ideas until you can get financing to get closer to the big time. + The Big-Time Competition: Should you worry? +.# Invasion of the Federal Government The Feds helped make the Internet the incredible success it is today, so it seems sensible at first blush to suggest that this should continue. As you probably know, the government is in fact abandoning the Information Superhighway, heading for the exits just as the place is starting to look pretty darn impressive. Many people have in fact wondered out loud why the government doesn't just buy a bigger backbone and stay in the business. The Internet was a highly successful government program mainly because they didn't do much. Yes, they provided the funding, but the key to the Internet's success was that they didn't put very tight control over what it was used for. Thus the curious fact that a sizable percentage of network traffic is alt.binaries.pictures. erotica.female instead something sensible like scientific reports. The Internet has been operated not by the government, but by local sites. Unfortunately, this non-governmental control combined by Federal funding just cannot last. Now that the "Information Superhighway" is becoming a more prominent part of people's lives, it's only a matter of time before Fundementalist Christian groups try and get alt.sex.stories booted off the net. And if our woozily incompetent government is still in control, there's a pretty darn good chance of that happening. Best of all, the lack of government control over the net gives us the ability to start our systems, run them as well as we can, and even have a little fun doing it. Don't ask for the government to return, or we might get the Post Office of Internet providers - slow, expensive and stupid. +.# Invasion of the Phone and Cable People Cable TV companies represent a more interesting competitive threat. For just $ 75 or so a month, they tell us, you can have a circa 56k connection to the Internet, providing you with far higher quality fthen you'd ever get through a traditional ISP. PSI recently conducted a joint experiment with a cable company in Cambridge, MA. Obviously this is a very biased location because more computer and Internet lovers live there than virtually any other place; you would expect firm demand here even if the whole thing was a disaster everywhere else. (True, many people would have access through their places of employment. Still, there are more people who would find the sheer technological "win" of interest than anywhere else I can think of). I have read that the venture was a disaster, with hardly any signups. My suspicion is that few people wanted to pay those prices, when they could get a $ 20/month account from an ISP. However, it's also possible that people are using their free university and work-based Internet accounts instead of going with PSI. Many Internet users, however, normally use separate accounts for work and home, so my guess is that the PSI service was just too expensive for the benefits offered. | Update: PSI's Cambridge venture into Internet over cable TV has | apparently been cancelled, per Karl Denninger . I received a correction on this long-standing part of the FAQ from Jeffrey Shapard, who apparently worked on the project: -- I was a bit amused by the commentary on our PSICable project up in Cambridge, Mass, where we did early work on delivery of Internet service over a CATV plant. You refer to it as a disaster. Actually, it never went to market. We actually considered it a success, in that we managed to get early technology working well despite the wild hairy beast of a live 2-way cable plant. But we also figured that the difference between what it would cost to provide the service well versus what folks would actually be willing to pay, given all those less sexy but tried and true alternatives, just did not make a good business case. So, with tears in our techies' eyes (mine, too!), we put it on the shelf for another day and moved on. -- Before ending our discussion of cable TV companies, it's worth noting the results of another venture, "video on demand". This was supposed to be the holy grail of cable TV, the service that would pay all the fantastic costs associated with the "information superhighway". A distinctly low-tech test was created, where a rack of VCRs was hooked up and a person hired to grab the requested tapes and load them. Prices ranged from $ 0.99 to $ 4.00 per view. For a typical video on demand venture to be solidly profitable, executives predicted that roughly four rentals per month would have to be made per customer. The actual figure was slightly under two, attained with the kind of promotion blitz that couldn't possibly be duplicated on a wide scale. Apparently people actually enjoy making the trip to the video store to check out their evening's entertainment. (This was reported in a recent issue of Wired (I think September 1994) and in the LA Weekly). My conclusion from these two points of information is that the telephone and cable-driven superstructure is likely to be a flop. Internet services would require far more complex connections than video on demand or home shopping, and I doubt that the profit potential is as high. As a result, I doubt that we have much to fear from the "cabledroids". Phone companies, such as Pacific Bell, have gotten into the Internet business, but despite reasonable rates, they don't seem to have much effected the ISP scene, although cross-selling with second phone lines might be a trend to watch for. There is, however, one thing to watch out for: Many telephone companies are arguing to drop the traditional local calling areas and replace it with metered service. If they did that, and offered an unmetered Internet connection, they could own the market. Be sure to do your darndest to make sure this doesn't happen in your state! ISDN might be another version of this, which should in theory be able to offer switched 56k connections to the masses. So far, the phone companies have been very sluggish to promote this service, but with the ever-increasing demand for bandwidth this is bound to change in time. Unfortunately, such an event is bound to hurt the typical ISP big-time, since the T1 lines we buy (forget about a 56k when your customers use ISDN!) will carry about 1/3 the users they did previously. Whether the public is willing to pay the increased costs associated with such a service is open to question, considering the failure of early cable TV efforts discussed above. ISDN has the potential to be a sneaky way for the phone companies to get back into the measured service system they know and love. It looks like they have lost the battle for measured service for voice phone connections, but ISDN is a brand new game. Right now, Pacific Bell makes residential ISDN available on a measured basis during the day, and a non-measured one on evenings and weekends. We'll see if people take the bait. Pacific Bell, however, has recently doubled its ISDN per-minute charges, which makes me think the phone company is really rather clueless in its marketing; I'd have a hard time recommending ISDN to anyone not using the connection for heavy business use in view of these much higher charges. Karl Denninger offers ISDN on his provider. When I asked him how well it was working, and how good customer acceptance was, he said, "It's dogs---; interoperabilty problems galore." In short: Keep your eyes open, but at this point I don't think the competition is likely to be as bad as it looks. +.# What about public libraries and Freenets? Well, this is a strange one. Here in California, we hear about libraries shutting down for lack of support every day, and yet people are talking about hooking them up to the Internet! I guess Pac Bell will be donating the service or something, since otherwise that would look like a pipe dream. There are two schools of thought among ISPs when it comes to library connections to the Internet: (1) They are evil, government-subsidized organizations who might become ISP competitors "through the back door", giving free or very inexpensive services. (2) Why not join them, not beat them? Most libraries really don't have the resources to run themselves, much less operate an ISP. Give them space on your ISP, and let people have limited free access from library-based terminals. Then, if they want more, or if they want modem-based access, they will naturally come to you first. However much I may dislike government-based organizations on the main, I still think (2) is by far the best thing for an ISP with a potential library competitor to do. Freenets generally offer very limited Internet access. A large number of my system's customers have come from the Los Angeles Free-Net; they don't care for its censorship policies (which I gather exclude the sexual stuff) nor for its very limited range of systems you can telnet to. As a result, I think you could consider your local Freenet a nice way to introduce people to the Internet who you can then talk into becoming your customers. Frank Hecker was kind enough to write with some additional comments on this issue. Frank is on the board of directors of CapAccess, a Washington, DC community network; he was also its "administrator and chief technical honcho" for its first couple of years. ("Community Network" is the public domain term for "Free-Net"; the latter is a trademark of the National Public Telecomputing Network. CapAccess is not part of NPTN and therefore not a "free-net", even though the same general concepts are used). Not surprisingly, Frank believes that your local library or freenet can be an ally, not an enemy. "I believe strongly that community networks and commercial providers have many potential areas of cooperation, and are far better off in the long run cooperating than competing." ISPs tend to have the technical ability to set up a complex networking system; community networks and libraries have sources of information and close ties to the community. "My position has always been that community networks should encourage users to move on to commercial providers as soon as possible, especially if the users' main interest is in Internet access as opposed to community information (which is the raison d'etre for many community networks)." Frank suggests that the future of community networks is less in providing net access and more in providing information to the net. Competition has brought the Internet to the people at an increasingly low cost, and community networks don't have the time or fiscal strength to keep up with net administration tasks. It might make more sense, then, to set up local newsgroups on the community's commercial Internet provider, and use it to run Gopher and WWW servers. Even if the community wants to run their own server, confining it to local content and the provision of information through the WWW might make better use out of scarce community resources. (Just a full USENET newsfeed takes about 130MB a day, so a few days' of global news would tax the type of equipment a community provider could normally afford). The only problem with this vision is that people who don't own computers, or those who are too poor to afford an ISP's charges are effectively frozen out. Libraries are particulary interested in this problem, as the traditional providers of information for people with intellectual curiosity but no money. Because of this, Frank says: "I found your suggestion about ISPs cooperating with libraries to provide limited free access a particularly useful one in that regard." "The bottom line is that I would strongly recommend that any small local ISP get to know the local people involved in community network and "Free-Net" activities, and see if there are any possibilities for joint ventures. Going into this, you should recognize that their attitudes and motivations may be significantly different than yours (especially if you're a net.libertarian) and that in many cases they will have emotional and institutional reasons for wanting to run their own systems. (Generally the larger the institution the stronger these reasons will be, which is why ISPs in small communities may have more success with this approach.) But in all cases I think ISPs will be better off going into discussions with a "win-win" attitude as opposed to demonizing community network people as proponents of public give-aways and enemies of the free market." In late 1996, in a move that was symbolic of the problems of the Freenet movement, the umbrella Freenet organization went into bankruptcy. The text-based model of the Freenets with its clunky software was just not what users wanted. +.# Microsoft and AT&T, the Terrible Two The hottest news in November 1996 was America Online (AOL)'s conversion to $ 19.95 a month flat rate. Considering their poor profitability at their previous higher price points, it's unclear as to how they'll be able to afford the results of this announcement, even if subscriber churn goes down significantly. Most ISPs who have talked to AOL users say that Internet service quality is a far more important issue to them than price. If this is so, the worsening user to line ratios on AOL is likely to cause severe difficulties in the near future. Because of this, most ISPs are not worried, despite the audacious nature of AOL's move. (Now, in July 1997, we know this is exactly what happened; AOL got a massive black eye due to connection problems the ISP community predicted as soon as the changes were announced). AT&T has put forth a major threat to existing ISPs, with their offer of $ 19.95 per month for unlimited access. Early reports have support as being quite poor, but nobody underrates the threat of a major national name getting into the business with highly aggressive pricing. Note that this rate matches some Internet providers, but doesn't beat them. I think that, if you're offering high-quality service, people will continue to use you and recommend you. Many providers feel that the influx of customers generated by massive advertising campaigns will be good for their companies. These national organizations appear incapable of providing high-quality service; all those who have tried so far have failed, including AT&T and MCI. Our good friend Bill Gates, of course, wishes to take over the entire world of computing, so consumers will soon be using Microsoft Windows to use a Microsoft online service to hear more about Microsoft products. Those who to hear about IBM, Lotus or Novell will probably want to use the Internet, so of course Microsoft will provide that, too. (I wonder if the new service will use Windows NT SMTP, which is still known in the industry as a bug-ridden product). The Microsoft Network's evolution over the year or so of its existance has been rather intriguing. They never got much of a content base, and as a result they decided to move into primarily offering Internet access. Since their rates are higher than regular Internet providers, I don't think they'll have much impact on our core customers, except for people who need a nationwide dial-up network. One fascinating fact, reported in the Wall Street Journal, is that Microsoft has now basically deserted their content providers in their rush to the Internet. As a result, the system described in the paragraph below may not even exist anymore, but I thought the concept was interesting enough to retain. Microsoft's new Microsoft network (aka Marvel) is meant to bring the online service world into a new age. In a hype-filled press release available via Microsoft's web server, they made it look like death time for all Internet providers and online services, however well run. A base fee of only $ 4.95 a month, combined with a de-emphasis on hourly rates, is supposed to make the Microsoft Network more affordable and higher quality than its rivals. In an interesting innovation, the cost of running the network was meant to be borne by the content providers, not Microsoft or the users. For example, let's say Time magazine wanted to get on the Microsoft Network. They would have to pay Microsoft's internal fee for use of the service, and they could raise the money by (1) charging their subscribers, or (2) charging advertisers. So if MS's internal fee is $ 1/hour, Time could charge subscribers $ 1/hour billed via MS, or they could sell $ 1 worth of targeted advertising during that hour, or they could charge $ 0.50/hour to the consumer and sell $ 0.50 of targeted advertising. And, of course, they could charge higher rates than this (say $ 2/hour) and make money. How Time would raise the money was completely up to them, and they could bill any amount they pleased for any service. So if Lexis came on to MS Net and wanted to bill $ 50/hour, they could. What this means to the Internet is most mysterious, since there is no real "content provider" who could be charged. My best guess is that UUNET becomes the content provider (since they are the people running the MS Network connection). If they have to charge $ 1-2/hour for the service, as they almost certainly do, I don't think we have much to fear from them. I don't think the type of consumer who likes the Internet will be too fond of Microsoft's offerings, which I feel are likely to be pretty well sanitized. So in sum, I doubt that the MS Network is likely to be the disaster for Internet providers it's been considered. If anyone has additional information on this topic, though, I'd enjoy hearing it and would update the FAQ accordingly. Recently, Microsoft's entire strategy for their MS Network has been changed to basically fold it into the Internet instead of fighting it. This seems to be a substantial retreat from their earlier efforts; what they will do with MSN itself is hard to say. Stay tuned. The good news on the Microsoft side is that their PPP support is genuinely well regarded, far superior to anything offered with Windows 3.1. With the growth of Windows95, it looks like offering Internet accounts will be far less painful in the future. +.# America Online (AOL) For the longest time, AOL seemed to be pretty laughable as a competitor. Not only was their service terrible, it was also way overpriced compared to an ISP's. Their service is still bad and likely to get worse, but in a recent bold move that I suspect will cost them dearly, they have matched ISP pricing - for $ 19.95/month, you can get all the AOL you can stand. If, that is, you can get online. I predict steadily lower revenues and more busy signals for the online service, which should send plenty of new customers our way. AOL also made some key mistakes, such as putting all customers on the higher cost plan. They expected this to help offset the loss in revenues from their high-end users, but it backfired in very bad PR for them. Some providers have noted that AOL's moves have had a negative impact on customer acquisition. I'd like to hear from more people when this issue stabilizes, since it's quite possible that things will shift dramatically in the near future due to service problems with AOL. They were bad before; with the unlimited policy, they can only get worse. + Equipment +.# What are your bottom-line recommendations for ISP hardware? This depends on exactly where you are on the learning curve and what you want to do with your equipment. If you're not too familiar with Unix, or if you just want the coolest equipment on the block, I recommend Silicon Graphics (SGI). I use SGI equipment, and I continue to recommend it because it's the most enjoyable Unix to use. The main disadvantage is a lack of commercial software and high compiler prices (GCC is hard to get running). A good starter system is a SGI O2 with 192MB RAM and an external (third-party) 9GB disk. That will work great as a mail and web server. However, I don't recommend the WebForce bundle because you can buy a Macintosh + software for about the same as the SGI software alone, and it won't burden your main machine. Support is excellent and reasonably priced (Sun's support is a lot more expensive, and I don't hear too much good about it). Greg Douglas at http://www.reputable.com is an excellent dealer of used systems that I recommend without reservation. To run a solid, professional operation, I recommend Sun UltraSPARC equipment. Sun has a reputation for being solid, and the public domain software most of us wind up using compiles easily. HP/UX is also worth a look due to the extremely high quality of HP equipment; however, it's known as a mutant version of Unix; that makes it hard to get software running. I recommend getting Sun clones instead of Suns; Ceram at http://www.ceram.com sold me the Sun clone I have running www.amazing.com even to this day. It's been my web server for years and years and has never given me a lick of trouble. If you're very familiar with PC hardware, or if you have a small-scale system and don't have the budget to do things the more professional way, I recommend Linux. It's cheap and available on pretty much all PC hardware. If you're not familiar with PC hardware and want to go the cheap route, I think the Apple Macintosh is worth a look. It has a small but devoted group of fans, and it's more secure than anything else out there. +.# What kind of computer(s) do I need to become an ISP? I don't know a frightful lot about hardware, but this section of the FAQ has stood the test of time and comments pretty well. The main exception is a great deal more information from BSDI fans about their favourite system. In addition, as I have been actively shopping for Sun hardware, I've talked extensively with both a Sun reseller and a reseller of Sun clones. The occasional Linuxer has made his presence known as well. If you have some comments on this information, please speak up; I'd like to be able to flesh this section out a bit more. To start with, you need some computer capable of running Unix. Opinions vary dramatically over what operating environment is best, but few who have been in this business long use anything but a Unix derivative operating system. Here are a few capsule arguments for various versions of Unix; corrections are welcome. Remember, the most ferocious holy wars are often between those of strikingly similar beliefs! Christianity versus Judiaism, Sun versus BSDI versus Linux. Watch the flames grow! +.# A Summary of this Section with Supporting Documents Recently, another Sun versus PC debate broke out on the Inet-Access mailing list, which was unusually enlightening. Here's an executive summary; in the HTML version of this document, I have attached pointers to two messages, one from a Sun lover and the other from a PC enthusiast. I hope this will help interested people make a decision. On the Sun side: Advantages: - The most stable solution, easiest to set up - Good hardware decisions are made for you already; you don't need a hardware expert to figure out what to buy. - Most of the public domain Unix software is easier to compile on a Sun than any other platform. Disadvantages: - The proprietary components are harder to locate; you can't just drop by your neighborhood PC store and buy a replacement if something goes wrong. Note, however, that many components, such as memory and SCSI disk drives, are standard. On the PC side: Advantages: - You can get a slightly better machine per unit of price. (The difference is probably less than you think). - You can exchange parts with your existing PCs, if you have any; if your ISP business fails or you decide you don't like it, you can wipe Unix off the system and use it for DOS/Windows. Disadvantages: - You need to know something about hardware to get a system that will work. - Setup can be very painful (it was for me). +.# Sun Workstations and SunOS/Solaris If you have a ton of money to blow, go right over to your local Sun distributer and pick up a couple of nice little Sun Ultras. They're just a shade pricey, but net wisdom has it that you'll have the least trouble if you go this route. Sun owners are generally quite happy with their technology, but some admit that the PC Unix route is so much cheaper that the entire world is going in that direction. | Or is it? Sun clones, surprisingly enough, are not too different | in price from Pentium PCs, once items included in the former are | added to the latter. For example, every Sun clone system comes | with Ethernet and SCSI at no additional charge. See the section | on Sun hardware, below, before writing off Sun as too expensive | for your application. A short word on a very emotional topic: Many people have run Suns for years, and on the main Sun has rewarded their loyalty with high-quality and much loved machines. However, a few years ago, a dreadful event happened: Their beloved Berkeley based SunOS was effectively replaced with Solaris, a somewhat slow and bloated System V OS. Because of this, just about everything that ran on a Sun had to be extensively rewritten. Many people stuck with SunOS because they didn't want to rewrite their software, they didn't want to buy new versions of their software, and they enjoyed the superior performance of the old system. In addition, early adapters to Solaris were confronted with a baffling series of bugs, problems and midstream changes that eroded their loyalty to Sun. Over the several years Solaris has been out, the bugs have been fixed, performance has been cleaned up, and all in all it's said to be a nice OS. If, of course, you can forget Sun's betrayal of a bunch of formerly happy customers. It is this and not any (or at least not many) intrinsic failures of the system that causes Solaris to be mentioned so negatively in any discussion of Suns. If you want a new Sun, and in particular if you want to take advantage of the new multi-processor architectures, you need Solaris; SunOS effectively does not support more than one processor. None of this should prevent people from buying Sun hardware, which is apparently still quite nice stuff. It remains the standard for running an Internet system. (Much of this was contributed by Scott Hinnrichs ). A good compromise was recently suggested by the aforementioned Craig Warner of . Instead of getting an expensive SS20 with multiple processors, pick up a couple of SS5s with a single 85MHZ processor instead. That system roughly equals the performance of a single processor SS20, at only a shade over half the cost. By buying such a system, you can stick with the tried and true SunOS. This compromise is no longer recommended, for a simple reason. As Sun hardware is updated through their Ultra series of chips, you may have no choice but to run Solaris in order to take advantage of their higher performance. SunOS is now almost completely phased out, much to the disgust of other users. +.# Silicon Graphics (SGI) Workstations SGI is the Mercedes-Benz SL600 of workstations. Classy. Expensive. Slick. Those that use 'em, love 'em. Those that can't afford 'em, hate the fact that they can't afford 'em. Based on posts I've seen in the misc.forsale.computers.workstation newsgroup, they seem to have the highest resale value of any computer I've seen. In October of 1996, the SGI O2 and Origin 200/2000 series systems were introduced. Early reports are that these are outstanding systems which will serve ISPs well. In fact, these are the systems I am using for my own ISP project. People who actually use them as web servers think they're great. People who use them to develop web site graphics love 'em. It might not be a frightfully good idea to use a SGI box as your shell machine, however; the default SGI configuration apparently disregards security almost completely. That may not matter too much if you're just putting up a non-secure Web server; it seems to me that SGI might be a first-class web server + web site design machine, with Suns or PCs doing the grunt work of servicing shell accounts and news. After all, do you really want users playing around with your Mercedes-Benz? A quick net surf showed that SGI has one of the best net presences I've seen. Friendly SGI employees answered most of the questions on the SGI newsgroups I checked; the SGI FAQs are some of the best I've seen on any subject; the WWW site is slick and inviting. Many of the users who asked questions on the SGI newsgroups showed a fierce affection for their systems; this was in sharp contrast to the Sun groups. SGI used to be pretty tight-lipped about its pricing, but now pricing for the low to mid-range (O2 to Octane) systems is available on their web site. You'll have to register, but it's a fairly painless process. Pricing for complete systems is actually fairly reasonable; pricing for components isn't. I have a WebForce O2, but I think you can get most of what you really need from an O2 from a non-Webforce version. The version of Photoshop they include is old and has mind-bendingly massive memory requirements. They do have some cool tools to create things like buttons, and I'm glad I have it, but for the $ 1,000-odd cost, I think I would have been better off getting a Macintosh with the latest version of the graphics programs. SGI has many plus points for the look and feel of the environment and the sleekness of the GUI, especially for users who would otherwise consider NT systems for their ease of use. SGI is easier to use, and it's also deeper if you want to probe inside and really understand things. Absolutely incredible, mind-blowing graphics software, such as Alias|Wavefront, is available, but it's not something you want to give your graphics tyro - their entry level package appears to cost about $ 7,000. Unfortunately, other software, such as mid-range commercial databases, is hard to find. However, the excellent shareware mySQL package [ http://www.tcx.se ] has filled this gap very well, with performance comparable to or better than costly solutions such as Oracle. I've found support to be very helpful so far, but I haven't had much need for it. Service/support contract prices are reasonable compared to, say, Sun's. The opposing view comes from Mark Tempest , who says: "If you plan on having your SGI workstation on the Net, be prepared for someone to spend lots of time on making such a host secure. Very secure if you plan on using it as a secure server. SGI, in recent threads on comp.security.unix, has taken quite a bashing over their stance on security issues surrounding their workstations. One SGI employee, while I don't think he spoke for the company (in fact I'm pretty sure he disclaimed it) was heard to say that the machines come configured for internal use in a lax security environment such as a corporate LAN, and not as a plug-and-play node on the Internet." The SGI Administration FAQ has a commendably comprehensive list of known SGI security holes and how to fix them; I was quite impressed by its thoroughness. In fact, the whole SGI series of FAQs impressed me a great deal in terms of near-obsessive completeness. So you may not want to run credit card numbers or digicash through a SGI system. If you want to do that, you should get a separate Apple Macintosh server that does nothing else; unless you're an obsessive security professional, no Unix or NT system is secure enough for that. Because the Mac wasn't really designed for remote access, it's very hard to crack. I think the O2 has cleaned up a lot of SGI's act on security - there are some very nice utilities now to automatically disable most features that can introduce vunerabilities. As with any vendor, it's best to replace many of the vendor-supplied software (especially sendmail) with newer versions or alternatives. Steve Davies has another type of cautionary tale. "Our experiences with SGI equipment have not been so trouble free. We have several SGI machines, Indy's, 340's, etc. We used them to replace some older HP boxes (HP 3000s), mainly for client/server database applications. We have had several CPU and disk failures over the last couple of months, whereas the HP platforms rarely skipped a beat (one disk failure in many years, even that event was well anticipated)." Note that SGI is System V, not BSD. This means that it may be somewhat more difficult to get some networking programs to run, just as this is also true of Sun's Solaris (see the detailed discussion of this in the Sun chapter). If you want to buy a workstation because you have money to burn, and would just love to do all sorts of neat things with it, I suspect the $ 6,000+ SGI O2 is the ideal machine for you. For why I think this, check out their web site, http://www.sgi.com/ . Finally, If you're a snob, you gotta love SGI. Their marketing people have brains, a real rarity in this business. Your Web site can have a neat "SGI Powered" logo that you can use to tell people your Web server runs on computers that are more expensive than God. My systems have now been running since April without a single hardware failure or serious problem. +.# PCs running BSDI Unix A small group of people on the Inet-Access mailing list have recently given BSDI poor marks in support. Other BSDI users, however, have responded with loyalty, saying support is still fine. Karl Denninger, a traditionally strong supporter of BSDI, turned against them when they refused to give him priority service when he promised to buy a support contract. Note, of course, that this means he did not have one at the time! There are still many BSDI loyalists, although some have still questioned the now much higher cost of source code. Before that, here's what Karl and others had said about BSDI. It should be interesting to see what happens in the future. BSDI users are ferociously loyal to their system. Karl Denninger, probably the most successful provider on the Inet-Access mailing list, uses modified BSDI systems with, if my memory serves, 64MB of RAM and a 1.0GB hard disk on each. (Karl has not corrected my memory, and I'm sure he's seen a copy or two of this FAQ). Each one can service approximately 64 users when a terminal server is used. He has told the world that the system is very solid and technical support is superb - a rarity among operating systems, or any other software for that matter. Source code license is $ 995.00; binary is $ 545. Once you buy either initial license, a license for each additional machine costs $ 250. Second-day Fedex shipping is included at these prices. I believe they are willing to negotiate a site license for very large numbers of machines. Karl Denninger will probably tell you you need the source; he's modified it extensively. In addition, Eric Raymond said, "Pay the $ 995. it's worth every penny", and Mark E Mallett says, "I agree with the support for the system." You can find out more about BSDI by mailing to info@bsdi.com. The ability to use a BSDI system as a router may make BSDI the system of choice for many providers; this could save you about $ 1,300 or more, depending on configuration. So you could buy a BSDI license for the money saved from the router alone. However, you should read our section on Routers below before making that decision. One possible dark spot in using BSDI is that there are fewer drivers available for it. According to Mark Tempest , most card makers in the PC Unix market are concentrating on Linux and SCO. Card makers are willing to send BSDI their cards, but BSDI doesn't have the resources to create drivers for many of them. This is one area where the distributed nature of Linux development is a major plus; as long as someone, somewhere has the card, a Linux driver is probably being written. I have heard from someone (whose name I unfortunately forgot) that BSDI is trying to wean its users off source code. In my opinion, this is a mistake that's likely to turn people to competing systems with source available, such as FreeBSD or Linux. There is a pre-set product called the BSDI Internet Gateway Server which is available in a 16-user license for around $ 995. I'm not sure how different it is from the basic system, but it sounds like prices may have gone up significantly since I last looked. +.# PCs running Linux I've revisited Linux recently because I finally decided to take Windows off my IBM ThinkPad and replace it with Linux. So far, it's worked quite smoothly, but I haven't used it for anything really important. I don't think there's any question that it will work fine as a web server nowadays. The main deficiency with Linux has to do with PC hardware, which just isn't designed to take large amounts of stress. Your $ 2,000 PC may be an ok machine, but it's not designed to be on all the time with its disk grinding like crazy as a typical workstation is. Once you buy a machine really designed to take it, you'll find the price surprisingly similar to "real" workstations like Sun or SGI. Linux seems to be the number one choice among providers coming up from the BBS world. It could be thought of as the latest and best continuation of the "Hacker Ethic", the belief that software should be free, and people should get the source and play around with it. (For information on the "Hacker Ethic", see Eric Raymond's 'The New Hacker's Dictionary' [Second Edition]. It makes me nostalgic for the years of my adolescence, as a ITS PDP-10 hacker at MIT. See the appropriate entries for details). The lack of $ 1,000 licensing fees for the source code probably has something to do with Linux' success over BSDI, as does the notorious cheapness of computer hackers. Support, surprisingly enough, is excellent. Post a question on the newsgroup, and you'll get friendly answers with good information within hours. Although the networking code is infamous for problems, I've had surprisingly little trouble with it. My present system networks a 85mhz Sun clone with my Linux PC; the Linux PC is connected to the Internet through a 28.8kbps SLIP connection. Although it wasn't frightfully easy to connect, everything is now working surprisingly well, with little trouble. Even after over a year of operation, the system has successfully withstood quite heavy loads. The main problem I've had is that the system occasionally (once every two weeks or so) crashed due to SCSI timeouts. However, this apparently was not a Linux problem; aiming a large fan directly at my drives appears to have solved it completely. Under the new regime, the system stayed up for 47 days without crashing, a considerable improvement. The Debian distribution is the most recommended one for ISPs, with Red Hat also getting good marks. The Debian advantage is mainly in better support for updates, both security-related and others. If you get Linux, get on the big-linux@netspace.org mailing list (I think email to big-linux-request@netspace.org will do this). You should also check out the linuxisp list; send mail to linuxisp-request@lightning.com to do so. People on these lists will be able to tell you what kernel versions are stable. Depending on which version you get, you may be anything from very happy with your system to ready to throw it out the window. Don't give up on Linux until you've tried a kernel others agree is stable. A FAQ on being a provider using Linux is available at http://www.anime.net/linuxisp/Linux-ISP-HOWTO.html . It has to be said that the OS works very well, and I'm quite impressed by it. I've heard that it's not good for a WWW server, however, and since I think Internet Marketing is going to be an important part of my business, I'm planning to move the WWW to a shiny new Sun clone box I just bought (from, predictably enough, craig.warner@ceram.com, as mentioned elsewhere in the FAQ). Because a multi-user Sun license is very expensive, and because it would basically require that I buy an (also expensive) terminal server, I am leaving my users on the Linux PC. That might change as I grow more lines, since the multi-port serial card approach is apparently not at all scalable for large numbers of users. The alternative to buying the Sun was to buy a Pentium/90 system and run Linux or BSDI on it. Once you add up all the "free" components included in your Sun, the price really isn't too different from that of a high-end Pentium/90 system. It's also likely to be more reliable, particulary when compared to Linux; many of the PCI bus systems are not quite there yet. Bryant Durrell tells us of one unfortunate aspect of Linux. "I'd recommend it for small providers, but you must make security one of your primary concerns, as it is not a system straight out of the box; less so than most Unixes. You will have to make sure you follow the various Linux newsgroups if for no other reason than to stay up to date with that issue." He would like to remind you to establish a shadow password file, which is covered in more detail later. Fortunately, a shadow password file is mainly important to those who are offering shell accounts; if you don't offer shell, you may not need to use one. +.# PCs running FreeBSD Jordan Hubbard , director of the FreeBSD effort, was kind enough to drop me a line with information on the status of FreeBSD, a BSD derivative for PCs. FreeBSD is a relatively new (about the same age as Linux) Unix variant whose design goal is to combine the stability of BSDI/Berkeley Unix with the free status of Linux. It has two major advantages over Linux: (1) the code has been developed using a more structured process, so it's likely to be more stable and have fewer bugs, and (2) it's based on the BSD standard (as is BSDI), which many Unix users feel more comfortable with. In an earlier section of this chapter, I covered the big and ugly fight between old guard Sun users and Sun Microsystems when the company switched its OS development to Solaris (System V). The differences between FreeBSD and Linux are similar, although Linux' diversion from the BSD path seems to be significantly less serious than Sun's. The main strength of FreeBSD has always been its networking code, which has been honed and refined over the years of development at Berkeley. FreeBSD continues this tradition thanks to a machine with T1 access provided for the development team by cdrom.com, their primary distributer. ftp.cdrom.com , that company's FTP server, gets massive numbers of hits and has been very reliable using FreeBSD. I attribute the lower popularity of FreeBSD to the following factors, virtually none of which have anything to do with the quality of the software: (1) Due to the centralized development model, there are fewer drivers available, and patches are slower to come out. As a potential user, you should balance this with the fact that FreeBSD's software is likely to be more reliable and contain fewer bugs, for the same reasons. (2) The libertarian ethos of the Linux development system, where just about anyone can offer a driver through a complex, decentralized network, appeals strongly to most users of the Internet. Linux is the first system created whose development was, for all practical purposes, developed based on the Internet model. (3) Although both groups have their internal politics and differences, the BSD camp seems to be more fractitious. I attribute this to the more centralized model, which ensures that some people are firmly excluded from participation; this is less likely to cause trouble under the Linux model. (4) Since FreeBSD is less well known, the newsgroups are far less popular. When I last visited the general FreeBSD newsgroup for user assistance (admittedly around a year ago), there were less than 10% the messages on the Linux group, and a disturbing number were about a rather nasty controversy within the development community. In short, the adventurous user of free software should use Linux; the more conservative user will be happy with FreeBSD. In addition, the lover of uncensored, unadulterated BSD will most likely be much happier with FreeBSD. A mailing list for FreeBSD users, freebsd-isp@freebsd.org, has been started. To subscribe, drop a line to majordomo@freebsd.org. +.# PCs running SCO (Santa Cruz Operation) Unix This Unix exists, and was basically the microcomputer standard for some time. It was originally based on Microsoft's Xenix product, although I think it's now a variant of Unix System V. It's quite pricey, but few who've tried it would pick it over Linux. One particulary important thing to remember about it is that only 64,000 i-nodes (files) are allowed on a file system, making it truly hopeless as a news server. Even SCO's support, which you might consider a major advantage of a commercial product, is rated as poor by those who've used it. The bottom line is that either BSDI or Linux would be better choices for a microcomputer Unix system. A mild update: SCO has apparently listened to some of the screams of its agonized users; they finally fixed that lack of inodes problem. SCO is apparently also one of the few systems that can support multiple processors, which is very good considering the low cost of dual Pentium machines. The other system that can do this is Sun's Solaris. There are unconfirmed reports of Linux support as well, but this is probably a ways away. According to Kevin Kadow, SCO recently purchased Unix System Labs from Novell, which gives them ownership of the complete sources to System V Release 4.2. It is not clear what they will do with it; "the USL football has had many turnovers in the last few years." +.# Other Unix Systems The IBM RS/6000 and HP workstation users have a small but vocal group of fans. However, again, I know little about them. A few people have pointed to DEC Alphas as the current king of workstation price/performance. However, the consensus seems to be that these systems are sufficiently different from SunOS or BSD to make installing networking software on them difficult. Particular venom has been reserved for IBM and HP Unix versions, which are particulary eccentric in many ways. +.# Macintoshes running special MacOS Software Tim O'Neill recommends a few new ISP resources. Check out how Digital Forest won their Mac ISP spurs at http://www.forest.net/advanced/isp.html . Also, check out http://www.macisp.com/ for more Mac ISP information and resources. Marcel Brown, mbrown@edwpub.com, has volenteered to update the state of the Macintosh state for us. I asked him a few questions, he gave me a few answers: David: Can an all-Mac ISP be done? It is possible to run a full range of Interet server software on the Mac. Obvioustly HTTP, but others include FTP, DNS, RADIUS Authentication, NNTP, SMTP, POP, NTP (Time), and TFTP. The various Mac implementations of Internet software are generally very good, and in some cases are considered to actually be better than other platforms' versions. Terminal servers are a whole different story. Your platform of choice should have no bearing on this decision. If you are serious about being an ISP, your terminal server will be something like an Ascend, Cisco, or Livingston box. I wouldn't set up an ISP, regardless of platform, with a computer hacked together to be a terminal server. David: Why set up a Mac ISP? What are the pros and cons? The biggest pro in my opinion is ease of use (which directly translates into time saved during setup and maintenance). I don't know about other ISPs, but I'd much rather focus my time on giving the best customer support possible, instead of figuring out arcane UNIX commands (or trying to make NT work correctly). With Macs, I know that any problems can generally be fixed very quickly, allowing the servers to do their job, and me do my job. It is literally child's play to anyone familiar with Internet serving (which is a different story altogether). Big pro #2 is cost. It's been proven that Mac Web Servers have a huge price/performance advantage over any other server. Sure, maybe a Mac Web server can't keep up with a big UNIX box, but you can buy 3 or 4 Macs and cluster them at a lower cost (and with much fewer headaches, not to mention better redundancy) than one UNIX box. Now while there haven't been any studies showing that other Mac servers besides Web servers show equal value vs. UNIX or NT, my experience shows that the Mac servers' ease of us makes them an extremely cost-effective solution. Going back to pro #1, the Mac's ease of use makes them much cheaper to support than a UNIX or NT box. In order to support a Mac server, you need to understand TCP/IP, general Internet theory, and (gasp!) the MacOS. In order to support a UNIX or NT box, you need to understand TCP/IP, general Internet theory, AND be well-versed in UNIX (or NT). Again, you can spend your time learning UNIX/NT, or you can spend your time more constructively (like supporting the customers that are paying your bills). Cons: The MacOS doesn't have pre-emptive multitasking or protected memory. However, this isn't quite as big of a deal as it seems, because the lack of these advanced features don't neccessarily make the MacOS less stable when running server software. Of course, it still would be nice to have those features. Certain software isn't availible for the MacOS. But then again, certain software isn't availible for UNIX or NT, either. Ummm ... that's about it. Really. David: What about NExT/Rhapsody? I'm glad you asked! Can you say UNIX power combined with the MacOS's ease of use? Can you say Nirvana? Anyone familiar with NeXT/OpenStep will tell you how great it is for many things relating to serving. Dell computer might even tell you how they use OpenStep for all thier big-time customers they always spout-off about in their commercials (or maybe not anymore as Microsoft is suppossedly clamping down on that). But anyway, Rhapsody is definitely going to make waves in the next couple of years, because it will clearly be the best platform for just about anything. Sure, you'll be able to run Rhapsody on Intel boxes, but why do that when you can get true RISC performance from the PowerPC processor used in the Mac today (not to mention real plug and play from the advanced Macintosh architechture)? I simply can't wait for Rhapsody, as it will effectively wipe out any cons to using the Macintosh as a server platform. [28 June 1998] Rhapsody is going to be folded into the new MacOS X, available sometime late in 1999. Basically, this will be the same basic system as Rhapsody, with most MacOS APIs added to it; this makes it absurdly easy to port MacOS programs, and thus a much more compelling solution for Mac developers. This should guarantee that the Mac will become a much stronger server platform in the future. Here's some older information on the Mac, by a previous contributer. Scott T Boyd, who wrote the earlier section on using MacOS software, has been flooded with questions about how to set up a MacOS-based ISP. Unfortunately, he doesn't consider himself qualified to answer these questions for us. :-( He writes: I wouldn't count Macs out as Internet servers. There's a very active mailing list (apple-internet-providers@abs.apple.com) with a lot of discussion from Apple-only or Apple-mostly or Apple-centric ISPs. The ease-of-use that Macs are well-known for has extended into the Internet server software arena. You simply can't beat how easily a MacHTTP server can be set up. Drop some HTML files into a folder, launch the application, and you're serving up web pages faster than you can say "Bite me unix!" ... I would be ever so grateful if you would remove my name from the FAQ on using a Mac for ISP'ing (and the reference to my web site www.montara.com) and instead point people to a much better resource than me, namely http://www.solutions.apple.com/ I guess sometimes my FAQ creates more publicity than people expect! Sorry, Scott. :-( I do wish we had something other than this rather chilly looking corporate link, though - can someone else educate us on these questions? +.# Macintoshes running BBS software. Considering the high cost of running on this platform, it's surprisingly popular. This is probably because it's the best-developed graphical system we have, and so people looking for a GUI think Macintosh first. At any rate, a couple of companies have obligingly created BBS software that runs on the Macintosh and - at least on paper - connects to the Internet. Like DOS-based software, you get very poor newsreading ability. One of these programs is called FirstClass, which includes a quite nicely done graphical interface client. Unfortunately, their mailer is notorious for disasterous behaviour when put on the Internet. For some time before Canter & Seigel drowned them out, strange mail problems from FirstClass hosts were a high-traffic topic on news.admin.misc. Even though they may have fixed the problems by now, I still cannot in good conscience recommend this software. NovaLink is another Macintosh BBS program. Marlene Zenker runs this software and has had some quite fascinating problems with it. Apparently the system cannot accept a full NNTP newsfeed, and it has been a terrible pain to set up from start to finish. At this stage, electronic mail and telnet work quite well, but there are no WWW or Gopher clients. Let's be fair, however: most people don't find it frightfully easy to set up a Unix site, either. Still, the problems of dealing with half-baked, underdeveloped software seem to be significant in the Macintosh world. I'm afraid my conclusion about Macintosh BBS software is strikingly similar to that about DOS software: It's not ready for prime time yet. Perhaps at some point it will become a realistic, high-quality alternative to Unix systems, but I'm not counting on it. +.# Microsoft Windows Software Can you spell C-R-A-S-H? I thought you could! I don't think anyone has tried to use MS Windows software (such as the Excalibur BBS) to run an Internet provider, or a multi-user site hooked up to the net. However, the high-powered demands of Internet hookups are not going to be frightfully friendly to the fragile Windows environment. In short: Good luck finding anything that works, and don't tell me I didn't warn you. +.* What about Microsoft Windows NT? The recent troubles at www.microsoft.com have shown a number of significant weaknesses in Windows NT - and, perhaps more than anything, the corporate culture supporting it. Thanks to a simple DNS error, one virtually anyone could have made, all but one of their 14 servers was inaccessible for four days. MS blamed this unfortunate error on excessive demand for their servers; then they blamed it on a hacker; finally, the word leaked out that they were having DNS problems. Microsoft itself never really admitted what was going on. But hacking aplenty has gone on on the Microsoft site, with many people taking the challenge of trying to publically embarass the company. Indeed, on their "Scalability Day" in May of 1997, it was impossible to get on their servers. This should demonstrate a few interesting points to the audience. * Easy to use, "click and drool" software doesn't prevent anyone from making really, really dumb mistakes. * If you rely on Microsoft to answer your questions and take responsibility for a significant part of your business, realize that they may not know anything either. * Spending money on hardware like a drunken sailer doesn't help. Microsoft's web server farm has 14 quad-Pentium servers with 512MB RAM and 50GB disk drives, for their main pages alone (not including their search and database server sites). It still hasn't kept their systems running with any degree of reliability. As I write, they've been down or shaky for the last couple of weeks. Microsoft claims this is because they are upgrading the server farm. Actually, their own web page says that they are just establishing new sites in Europe and Asia. This shouldn't affect their US site at all; it's just a cover story for their failings. * Don't believe Microsoft propaganda. They've been caught in outright lies numerous times. The record shows that many people have used NT and abandoned it in disgust for Unix. I don't know of anyone who's used Unix and abandoned it in disgust for NT. In short: Don't rely on NT for your systems. Unix is proven, has plenty of software available, and support is as close as your electronic mailbox and Inet-Access. If you are running dial-up access, you should be aware that the NT Remote Access Service (RAS) is flaky and horrifyingly expensive to license. I had previously written that there was a $ 100 per line fee to use RAS. However, this is apparently incorrect; it's $ 20 per line. This still means that a 32-line system would cost $ 809 for NT server plus $ 20 per line x 32, or $ 384. So the total would be over $ 1,000 for something that could be done free using Linux. The next paragraphs are my older thoughts on this subject. The overwhelming majority of Internet administrators do not believe NT is the operating system of choice for an Internet service provider. This is probably a rather kind description compared to what you'd actually hear from a Unix user confronted with a question about NT. It is a matter of record that Microsoft uses Windows NT as their FTP server, named (with more hope than sense, perhaps) "gowinnt". It is also a matter of record that Microsoft.com, their mail server, is a SCO Unix box. Admittedly, Microsoft Windows NT is not, perhaps, quite as unreliable as your copy of Microsoft Windows. Hopefully. In any case, if they can't do electronic mail through NT, when they have the strongest possible reason to do it, you probably can't, either. Beware. There are now a few hardy souls who have attempted to use NT as a provider operating system. Some have had surprising success, and others have felt the horror of dismal failure. I had a long dialogue with one person who succeeded with NT; if you're pro-NT, don't give up on this section until the end. Larry Ash comments on the wonder of Microsoft: "Don't do it [use NT as your server system for an Internet provider]. Microsoft likes to play fast and loose with the RFCs. I know an ISP that insisted on doing it over everyone's objections. They told me last week it cost them $ 10,000 before they pulled the plug and switched to Linux." When I asked permission to repeat those remarks in this FAQ, he made one additional comment: "For long-term print I would probably prefer toning down the accusation that Microsoft doesn't follow the RFCs, but up to this point it seems accurate. ... The individual I was talking to said he had received assurances from Microsoft that they would have a RFC compliant DNS system by the 2nd of Dec. On the 10th of January they were reevaluating their decision to *ever* produce and ship one." Jeremy Porter elaborates a little on this question. "Windows NT could NO WAY NO HOW service as an ISP server. It needs: Name server support, NNTP server software, multi-user support, e-mail support. ... I understand there is a WWW server for NT, but I also understand it sucks rocks." Michael Nelson responds by noting that NT servers are easier to set up than Unix systems. The existing FTP and WWW clients work well, and host systems will become increasingly less important as we all go to SLIP/PPP. The name server, he says, is under development. Michael Dothar (dothar@intersphere.com) writes us as follows: (Begin long exerpt) Been reading your faq again and thought I might offer some insight on the cons and cons..er....PROS and cons of using NT and/or 95. Web Servers: O'Reily's Website (based on Bob Denny's WinHttpd port of NCSA v1.4) runs quite well on NT. I had it running on a 16mb 486/66 under NT v3.50 for many months at 50k hits per day over a 56k line. CGI is easy to write using NT Perl, VBv3 (though I hate basic), or C. Win95 does not support this system nearly as well. I can see no reason to use it over NT if you insist on using a Windows based product. Netscape Commerce Server on NT, on the other hand, is pretty much bad news. I mean, it 'runs' and all but there are many problems. It is painfully obvious that Netscape's server product guys are Unix programmers. The recomended procedure for running NT Perl scripts under Netscape is very very insecure (I wrote a wrapper application called "run" to handle the job). The C/C++ works better, but far from the standards one would expect from a Unix box. There is a problem with the length of a URL in Netscape, too. It seems to ignore characters somewhere after the 150th character in a URL. This was a problem for us in some applications because of the need for encrypted data and state information on the URL. We have not bothered with NSAPI (Netscapes method for writing modules that link in with the actual server code instead of using external CGI) because of our lack of comfort with the system as a whole. In general terms, CGI under NT is a problem because of some basic problems with DOS. Yes, Dos. You can't use an equals sign ('=') in a URL from a console application (aka, Perl, Most C CGI, etc). This applies to Netscape and O'Reily. I have not used Questar's product enough to comment on it. I prefer my Sun running NCSA v1.5 as a web server. I will be moving most pages off of the Website/NT machine some time in the near future. FTP: I don't like NT FTP for a few reasons, most having to do with it using the NT user registry for non-anonymous users. For that reason I have used WFTPd from Texas Imperial Software (alun@texis.com) v2.02 (32bit) for many months. It serves light anonymous FTP duties, and allows me access to my NT/Win95 machines in my office. It works. No complaints. I have not used it in a high load situation, so I can not comment on it's viability for serving in a heavy load situation. DNS: I use FBLI's DNS Service product for backup DNS on my network (primary is provided by SunOS 4.1.4 named). It works well enough in this environment. It configures just like a traditional UNIX based name server. http://www.fbli.com/ is the company home page. Other: For the record, Win95 is MUCH more stable than v3.11. The key, I have found, is to install 95 in it's own directory. This goes agains MS's suggested route of installing over the old Windows directory. Stability is greatly increased if you install the system fresh this way instead of trying to upgrade an old Windows or WFW system. (end of long exerpt) Other people have mentioned the high quality and variety of Unix-based software as a major reason to continue using it. When new systems come out, such as the Netscape WWW server, they're developed for Unix first. This will probably continue to be true for a long time to come. But what about the good side of NT? Hasn't anyone managed to get it to work? One brave soul, Bill Landry (blandry@questar.com), writes as follows: "I can tell you that we have been providing full Internet services on nothing but NT servers for the last 5 months. I will admit that when we started out we had to run 2 Unix servers because we could not find sendmail, NNTP and DNS for NT. However, over the last 10 months we have worked with companies to either port these services to NT, or have companies that have written their own from scratch. I will admit that everything we are running is beta at this stage; however, I must also say that we have gotten nothing but compliments from our several hundred customers that have switched to our server from other service providers who are providing Unix solutions. "Now, this is not to say that NT is a "better" Internet server. But it is certainly close to being "as good" as an Internet server. We provide dial-up 56KB, T-1, ISDN, Web, FTP, DNS, NNTP, SMTP, POP3, etc, again, using nothing but Windows NT servers. ... "The fact of the matter is that after using Unix for those first few months of operation, NT has been a welcome relief. It is a far cry from the command line shell of the Unix environment. It has been much easier to implement, use, maintain and trouble shoot than the Unix machines we were using." He adds that technicians from all the companies that created their software, including Microsoft, have visited his site and given excellent support. However, he says that all is now running smoothly without them, and he does not feel that the ISP who dares to run NT applications would have much trouble. I'm still wary, to say the least. NT might be easier to set up if you're in isolation, without support from the mailing lists and newsgroups that will help with Unix problems. I'd be surprised if the informal support network for NT ever gets as good as the one available for Unix. "Official" Windows NT support from Microsoft is available at a cost of $ 150 per incident. NT users who have been burned by NT's problems don't seem to be frightfully keen on this policy. A hardy soul has written a FAQ on providing ISP services using Windows NT. However, it is currently being revised, so I have been requested not to provide the link. Before you all leap out and pick up a copy of NT as your Internet server, I would like to remind you of a few things and give you a very personal plea not to run Windows NT. If you believe, as I do, that Microsoft is the 1000 pound gorilla of the industry, and that most of what it does is not in the long-term interest of Internet service providers, you'd be well advised not to volentarily give market share to Microsoft by using their server products. Why help them out when they seek to destroy us? Maybe you can run NT. Maybe you can run an operating system created by a company that wants to kill off every online service but the Microsoft Network. Maybe you can run an operating system made by the same people who created the wretched mess that is Microsoft Windows. Please don't. [David's Internet ventures, including www.amazing.com and freelink.net, proudly run Silicon Graphics and SunOS servers]. +.# What about Microsoft Windows 95? How on earth does Microsoft get more and more coverage, even as its software crashes on desktops everywhere? Early returns seem to say that Windows 95 is either the greatest thing since sliced bread, or the worst piece of trash to come down the pike since its predecessor product, Windows 3.1. The one thing we can say for sure about Windows 95 is that nobody - and that's nobody at all, folks! - is neutral about it. It even made Doonesbury, with the irresistable lines: "What's wrong? Why's the new Windows balking?" "It's greedy. It's holding out for 16 megabytes RAM ..." "Hey ... It just turned on the printer! It's a complete list of its demands!" "Give no quarter. Hold it to the box specs." Give no quarter indeed. Let that be my advice to you as well. Until someone tells me otherwise, I will assume that it, like Windows 3.1, is totally unsuited for an ISP operating system. However, if someone tells me more about the product, I'll do my best to listen. O'Reilly & Associates currently sells a version of their Website web server software for Windows95. I tried their 60 day free trial version on my Windows95 system, a Pentium/75 with 16MB RAM. I attempted to process a web server request while compiling a program in Visual Basic 3.0 (16-bit). The web server ground to a halt during the compilation and started up only after the compile was done. This is completely unacceptable performance for any type of web server application. Microsoft claims that the new 32-bit applications will allow for multi-tasking; unfortunately, all the 32-bit applications I've tested are larger, slower and buggier than their 16-bit equivalents. During typical programming use, I have to reboot my Windows95 system several times a day, as it frequently runs out of memory and/or system resources. Based on this experience, I cannot recommend Windows95 in good conscience for ANY ISP operations. I believe Windows95 to be roughly as stable as Windows 3.1 - which means not stable at all. +.# What about IBM's OS/2? OS/2 is a lot like Windows NT in its support for Internet services. The main difference between Windows NT and OS/2 in this respect is that Microsoft has actively promoted the former as an Internet server, while IBM has been silent on the latter's virtues. Lori A Martin gives her impressions of OS/2 Internet servers: "I wouldn't run an Internet provider service on an OS/2 platform because a Unix platform is best suited for it. However, I would feel comfortable putting up our FTP site on an OS/2 machine, or running our Web site on one, or running a gopher off it." For OS/2 software, you can check ftp.cdrom.com:/pub/os2/network/tcpip for client and server applications. You'll find plenty of WWW servers, finger servers, gopher servers and FTP servers. There is even a version of INN available, although I didn't see it in this site. Dave Hughes helps people bring up small Internet providers (16 lines or less) running OS/2. He says that Linux and other Unix-based systems are too complex and hard to learn when compared to OS/2. However, he admits that this is still not the right solution for a system using a T1 connection with over 16 lines, which is where most "real" Internet providers are. Since he and a few others are actually using it with some success, it sounds like it's considerably more workable than the much-hyped NT. I promised the public that I would gracefully apologise for my earlier statement that there were no OS/2 Internet servers other than for WWW. Team OS/2, please accept my apologies for this most unfortunate libel of a fine operating system. Unfortunately, that doesn't make OS/2 a suitable machine for an Internet service provider. However, if you were considering Windows NT as a Web or FTP server system, you might want to seriously consider OS/2; an 8MB OS/2 system would probably work just as well for a Web server as a 16MB NT one. I suspect that setup isn't any more difficult than NT's, and you certainly get a more modern user interface. +.* What about disk space, networking and memory requirements? When you get past a certain number of users, you will need to network several systems together. This is, for example, the approach taken by Netcom, which now has 23 SparcStations, massively equipped. Netcom's well-known performance problems, discussed elsewhere in the FAQ, seem to disappear when they buy new machines, and then pop up again 2-3 months down the road. When I first signed on to Netcom in March, they had 13 machines. As of the time I'm writing this (14 March 1995), they have 23. Right now, Netcom performance isn't bad, but if it follows usual patterns, that won't last. To start out in a very small way, you need at least 1 GB of hard disk space. To store USENET news for any appreciable amount of time, you'll need many times that. I can store about three weeks' worth of ALT.* on the one 1.8GB Quantum hard disk I have dedicated to alt news. (This includes the binaries groups, however, which you might want to expire more quickly or not carry at all. I keep the binaries groups for two weeks and the remaining groups for 25 days). Jim Dixon (jdd@aiki.demon.co.uk) strongly recommends SCSI disk drives and controllers; he is, of course, correct. You will wind up being virtually forced into this route anyway, since most large hard disks are available only as SCSI devices. Eric Raymond adds: "We started out with IDE disks to save money. *Big* mistake; we had nothing but grief from the cheap IDE controllers we were using. Go SCSI from the beginning; the reliability and performance is more than worth the slight premium." Sean Shapira has another viewpoint. "Linux will support 4 IDE drives, and these are commonly available at 500MB, for a 2GB system. Beyond that, a wise provider might consider distributing the load to a second server anyway. Admittedly IDE is slower than SCSI, but I argue the trade off may be worthwhile." (I wrote back to him noting that there are now 1.0GB IDE drives; he doesn't recommend them because they are not yet as cost-effective as the smaller units). Jonathan Heiliger notes that these are EIDE (Extended IDE) instead of IDE drives; BSDI doesn't presently support them. So if you're a BSDI fan, you'll have to wait on this one. (He is, of course, correct, since only EIDE supports more than two drives on a single device). According to Tony Sanders , a patch has been created which allows use of extended IDE drives. If interested, contact support@bsdi.com for more information. marcln@on-ramp.ior.com was kind enough to drop me a line mentioning that my comments on EIDE were somewhat out of date. "EIDE has become the rage, and Western Digital offers a 1.2 gig HD 10ms for $ 399." I believe, however, that SCSI performance is still superior, and you can put more drives on a SCSI controller than an IDE. Kevin Kadow seconds the many comments saying that IDE is a bad idea for an Internet provider. He also reminds us that most Unix systems can't go over 2GB per filesystem; as a result, there is really little value in using drives larger than 2GB. (As a general rule, the more separate drives you use on your system, the faster it will be). Darrin Stadler (torin@daft.com) has dropped me a line with a new modest update to this section. It seems like disk drive prices have been dropping steadily as I wrote this. "... You quote someone as EIDE drives being cheap at $ 399 for 1 gig. Well, you can get a SCSI 1 gig for the mid four hundreds in Computer Shopper. Another big advantage is that if you have a PC and decide to move to Suns or SGIs, you get to take your drives with you." I checked the Fry's electronics ad in the latest issue of Microtimes, and it looks like the price difference on 1.0-1.2GB drives is around $ 125. Not bad. I presently have two 1.8GB Quantums. If I were to get a new drive now, I'd probably get Seagate Barracudas - but you have to watch out for cooling. Because INN wasn't designed to work well with a news spool split between machines, I'm pretty much stuck using large SCSI drives on one system, and I suspect most others will be too. For the record, after reading Karl Denninger's comments on PCI's unstability, I postponed my Pentium/90 PCI + 9GB drive purchase, which I wanted to do together; I eventually got a Sun clone system instead. I suspect either solution will work, but the first one appeals to me because it requires no skill in mucking around with hardware. According to Tony Sanders , "Things seem to be better nowadays wrt PCI. Buslogic finally seems to have stable firmware for the 946. It's probably worth trying so long as you go with a vendor that you know will let exchange it if things don't work." His hardware recommendations are included later in this FAQ, under "Tony Sanders' Recommended Equipment List". Netcom gives each user 5MB of disk space a month, and charges for average usage above that figure. So a gigabyte partition for users will only last for about 200 users, and that only if you strictly enforce quotas. SLIP/PPP accounts, which are likely to be the bulk of your offerings, don't require you to offer disk space on your system at all. Eric Raymond notes that you can get away with a lot less if you use a customized BBS-style interface. "At CCIL, most of our accounts go through a custom BBS interface; the users don't have shell access. They get along happily on 300k each. Whether you can get away with this or not depends strongly on what proportion of your users are programmers." Mark E Mallett notes that 300k "won't even hold a .newsrc file. (Mine right now is 645k)." I suspect the difference is that Eric's users may not read news in any great volume, thus keeping the number of .newsrc files down. Mark continues: "We impose quotas to help remind people to keep things clean. We initially started out with no quotas; our disk space ran out within the first couple of weeks. Added quotas and got at least 90% of it back." My own system runs a custom-written newsreader that doesn't have a .newsrc file; it works with a list of newsgroups the user has entered. I've found that most people have extremely small groups files; a typical user directory (which I just checked) takes around 6k. The main key to this performance is that I let people "browse" through all groups I carry without putting them in the groups file; only groups that have been read at least once go there. In Tin and other readers, every group's status goes into the .newsrc file even if I've never read it, since I want to be able to see all groups on the list (even if I never enter more than a very small fraction of them). I would figure on about 300MB for the operating system, 1GB for users and 2GB for news, meaning that your disk requirements should be around the 3.5-4GB mark. Eric Raymond (quoted previously) seconded this motion, only noting that it should support significantly more users than I wrote. I suspect the key to this is really FTP. I don't allow FTP yet on my system, and won't until I have written a customized version that automatically downloads files to the user's system. That should solve most of the common problems with disk space. [NB Has anyone done this yet? I'm having a hard time getting the time for this particular venture, and my users are just salivating (you can see it, honest! :-) ) for FTP ...] +.# I want a Sun, but I'm confused. What would be a good sample configuration? The major problem with Sun, when compared to PCs running BSDI, is that they're normally sold to consultants or major corporations who know what they're doing. You can easily find someone to tell you how to build a PC, but it's considerably harder to learn how to put a Sun together. Most of this information is from Craig Warner of Ceram, a company that sells Tatung Sun clones. I talked to both him and my local Sun reseller. For some reason, the latter kept on mumbling stuff about "client-server" and "Oracle server applications" and such. He seemed like a competent fellow, whom I could easily entrust my Oracle server to. However, he seemed quite puzzled when I said that I actually wanted to run and use Unix on my system, instead of controlling it remotely with "easy to use" Windows applications through my Novell LAN. (I didn't know I HAD a Novell LAN!) The effect was as though we were speaking a different language. Craig, on the other hand, struck me as a very nice, capable fellow who was sympathetic to the needs of a budding ISP. "A starter kit, which can grow to support over 1,000 customers, runs about $ 20k. This includes modems (16-24), terminal server and SS20 type server. We use a ratio of about $ 30-40 in capital per account. Most of this capital has a useful life of 24-36 months." Here's a detailed price breakdown of the Starter Kit: [Note: This is years old. I'd appreciate updates from any Sun dealers who happen to read this document] Base SS20 Clone with CPU $ 4,000 SS20 model 50; includes graphics 64MB RAM 2,750 4.2GB disk 2,000 17" Sony monitor 1,000 20" Sony monitor $ 1,650 Sun-Bootable CD ROM drive 500 Plexor double-speed Total base system 10,250 Compare this with other hardware prices 4mm tape backup 1,100 Exabyte 4mm DAT 28.8kbps Modems @ $ 250 each 4,000 28.8 external vfc, v34 Annex Terminal Server 5,000 32-port with software and cables Power backup, misc cables, etc 1,500 UPS, surge protectors, ethernet Total accessories 11,600 Total 21,850 This was a real-world system that was delivered to a client. "This site was up and running quite quickly, and should scale well to 2,000-5,000 users. Additional hardware will be required, but the upgrades should be smooth (add memory, processors, modems)." On-site service on the system would run about 1% of value per month, or around $ 218 for this system. This is for 8 hour/5 day week service. Surprisingly enough, when I got a quote on a Pentium system from a Linux specialist dealer, it came out to only about $ 1,000 less than the Sun base system cost listed above, for roughly comparable components. This system includes a right to use license for the OS. Upgrading to an unlimited license costs $ 660.00. One suggestion made by my potential backer is that it would make sense to buy a second machine to substitute for the main one if it broke down for any length of time. This seemed like a sound idea to me as well. Because you could use a used SS10 for this, it might not be as expensive as you might think - around $ 7k or so would do it. Craig Warner course, endorses this idea: "Yes, this is a good strategy. It surprises me how few ISPs have a backup machine, given the cost of unscheduled down time." Craig can be reached as craig.warner@ceram.com. His online catalogue can be accessed at http://www.ceram.com/ | Sidebar: RAID Disk Arrays, the Technology of the Future! Well, | maybe. I asked my Sun reseller about the Sun's really neat-sounding | something or other-100 disk array subsystem. It turns out that the | thing is so fantastically expensive as to be way out of sight: | $ 24,999 for 6GB and $ 59,000 or so for 30GB. Gulp. Maybe I don't | want one THAT badly. For some reason, Sun hardware FAQs are not stored on the usual RTFM sites. Instead, you can get them and other Sun information from ftp://thor.ece.uc.edu/pub/sun-faqs . +.# Relative Capabilities of various Sun models I asked Craig Warner (yep, him again!) about the relative merits of the various Sun models. "Generally, a MicroSPARCII Machine (SS5) can support 25-50 users as you describe, or perform specific functions (i.e. news machine, WWW server, etc). The SS20 machines can support up to 150 users with currently available CPUs. Since the CPUs are modular on these machines, newer CPUs (now in the pipeline) will increase this capacity - probably to around 250 users. "On almost all configurations, a ratio of approximately 2 MB of memory per simutaneous user is a good rule of thumb. Inadequate memory will cause the system to start virtual paging - which will quickly bring the system to a crawl. "As a good case study, the main interactive server at Clark Internet Services is a SS10 (the SS20's predecessor), with 2 60mhz SuperSparc processors (fastest available). The machine has 250MB RAM and supports a peak load of about 150 users, from a customer base of 3-4000. They use a similar configuration, but with less memory (128MB) for the news system." His Internet Starter kit (see above) is a SS20 with the 50mhz processor, about 20% slower than the 60mhz model mentioned above. Since this was written, he seems to have changed his mind slightly about the best machine to get for a provider. He now suggests getting a network of 85mhz SS5s (about the speed of a single, 50mhz SS20), which can be had for around $ 5,000-6,000 each. There's no real savings, since you'll still need about two of them to handle the 150 users a SS20 could. But the incremental cost to get started is lower, and the smaller load on the smaller machine seems to make performance more sprightly (probably due to the lower load on the local disks and other peripherals). +.# What sort of monitor should I get with my Sun? Right now, I'm looking at a drop-dead gorgeous picture, and all it cost me was $ 2,595.00. That's right: the final stage in my quest for the perfect monitor led me to a NEC XP21, a beautiful but hideously expensive piece of equipment. (In November 1996, you can get a XP21 for a little under $ 2,000. Ah, how things depreciate :-( ). The saga began when I wanted a 20-21" colour monitor for the new Sun clone system. My Sun vendor (aka Craig.Warner@ceram.com) recommended the standard 20" Sony, and I ordered it. When it got here, I learned that it had a .31 dot pitch, making the picture a little blurry. As a result, I returned it to him and decided to get a NEC, my favourite brand. The NEC XE21, an otherwise beautiful unit, won't work with Suns. It goes up to a maximum resolution of 1024x768, despite what you may have heard in some literature, and as a result it was not really Sun compatible. The proper Sun compatible unit, complete with BNC connectors, is the XP21. So how much better is the $ 2,595 XP21 monitor than the $ 1,600 Sony? In truth, unless you're as persnickity as I am, you probably won't notice much difference. One major advantage of the NEC, however, is that it has both BNC connectors and standard PC ones; a front panel switch lets you switch between them on the fly. So if you use both Suns and PCs for various applications, but don't need to view them at the same time, the NEC is truly your dream monitor. (I do Windows development on a laptop PC, so being able to use it at home with the big NEC is particulary nice). If you don't have that kind of money, there are a lot of 19" Hitachi monitors around that you can get very cheaply. Unfortunately, picture quality is often (usually) poor. In particular, the tiny type used by many Sun windows is just too blurred to read. I strongly recommend a 20" or above monitor for your Sun system (or any computer running X-Windows, for that matter). Any smaller and you'll be squinting like crazy. Pity they're so expensive. I noticed that X-Windows is so unmerciful on small monitor users that one ISP I know actually prefers using the Windows-based net browsing programs from a PC! Horrors! My Linux system, which doesn't run X, has a 15" NEC 4FG monitor, which works fine on the text console. Many people would say it's a bit of overkill, but it definitely soothes my eyes after a tiring day. As does the new XP21. And isn't that what a good monitor's for? +.# Care and feeding of disk drives A major problem that you may face with disks is overheating, especially if you put more than one in the same cramped PC case. People who have bought Seagate Barracuda drives are especially vunerable, since they run very hot due to their high speeds. However, even my twin 1.8GB Quantums suffer from overheating; the symptom is a "SCSI Timeout" error and the subsequent need to switch off the computer and wait about three hours before turning it back on. There are two somewhat contradictory recommendations: The first one is to open the case and get a conventional small fan blowing across the drives. This is the solution I'm presently using; my system has now been up for 35 days, when it would rarely hit 14 before. The other recommendation is to leave the case on, because it aids air circulation, but buy a small internal fan and place it near the drives. According to Kevin Kadow, Seagate and Maxtor drives are particulary prone to a problem where the platter lubricant "cooks" from the heat. The main symptom is that the drive works great - as long as you don't turn it off. If you turn it off and let it cool, the head becomes glued to the platter, and the drive will not be able to come up to speed when you power it back up. "If this occurs, a good whack on the side of the drive will either jog the head enough to get it flying, or rip the head off the carrier, destroying the drive and all its data. Either way, the problem is solved." +.# All About the Sun Netra This is an amazingly long section, so I'll give you the straight scoop in the first paragraph, and you can read more if you're curious. The Sun Netra is said to be a "plug and play" "Internet Server" system, designed to hook up DOS and Macintosh networks to the Internet. As a result, it looked intriguing to many ISPs and would-be ISPs struggling with their configuration files. The straight dope, however, is that Sun's own engineers say that the Netra is really meant only as an interface between the Internet and a large corporate internal network. As a result, it is completely unsuitable for use by an Internet provider. Full stop; that's it. Now for the details. Here's the scoop on the Sun Netra Internet Server as my Sun reseller explains it. You buy it and put it in your closet. You plug it in to your network and your Internet provider, switch it on, and a soft female voice tells you what it's doing. In about 10 minutes, it silently comes up and starts running. You can then close the closet door on it forever; it's controlled through Windows applications set up on a Novell LAN. If you want to provide Internet services to people outside your organization, it appears to be a non-starter. It might work, however, if all you want to do is set up a WWW server. Jamie Saker was kind enough to write alerting me of the Netra's deficiencies. After his text appeared in the FAQ, it was read by Mike H Geldner, Tactical Engineer for Sun Microsystems, who took the time to write a response. So that this FAQ doesn't balloon to a truly frightful length on this question alone, I will summarize what the two agreed on, and mention the areas of disagreement. The remarkable thing about this particular situation is that neither party disagreed a frightful lot; I think Sun wrote me on this issue mainly so I would remove some of the more hostile words from this document. Here is precisely what Mike Gelder of Sun wrote in his message about the target market for a Netra server: "Prospective buyers who are knowledgeable about Unix, the Internet, Domain Name Services, sendmail, POP3 and all the other parts-and-pieces that needs to be manually configured and maintained are probably better off buying a standard Sun SPARCstation 5 or 20 instead of a Netra i. They would find the Netra i limited in its flexibility as it pertains to a general computing role." This is a key paragraph. It's equivalent to "If you're an Internet Provider, don't buy a Netra; buy a SS5 or SS20 instead". In short, precisely what Jamie Saker and his fellow Netra critics have told us. Who would want the Netra, then? Someone who wants a "no fuss, no muss" connection of his users to the Internet, strictly as a client system. That is, people would use the standard Windows or Macintosh Internet utilities, such as Netscape, Hgopher or Microsoft Mail. Packets would flow through those PCs into the Netra and hence out into the Internet; the Netra would handle sendmail, POP, etc, How do the critics disagree with this? The good news, they say, is that the Netra has achieved its goals of being extremely easy to set up. You can set it, lock it in that closet and leave it forever untouched. The bad news is that it doesn't include Internet service servers such as WWW and Gopher. As Jaime Saker comments, "Sun's marketing has convinced users that they're getting a fully functional 'Internet Server'." According to Michael at Sun, this was a deliberate design decision: "At the time the Netra i was introduced, the licensing constraints and potential support problems associated with distributing public domain software were a great business concern. Utilities such as NCSA Mosaic and Netscape WWW viewers were in constant development. A decision was made NOT to ship utilities of this kind with the Netra i in its first release. As described above, the role of the Netra i did not mandate that a complement of Internet access tools should be provided with it. Thus the decision was made to keep it simple (in the first release) and supportable." He suggests that ISPs might want to provide pre-configured Netras to their LAN customers, such as companies getting 56k or T1 links with them. "Buyers who want quick out of the box access to their Internet provider get exactly that with Netra i. An Internet Service Provider reselling Netra i systems has the ability to ship a configuration floppy with the system that contains information specific to the installation site and service provider. Upon power-up, the system takes about two minutes to read the disk, configure itself accordingly, initialize the interfaces and begin acting as a local resource for DNS, POP3 mail clients, etc. If the system disk crashes, a similar operation with the CD ROM OS release and the same configuration floppy will rebuild and reinitialize the system (with a new hard disk) in about 30 minutes. The ease of configuration and servicability is what is being sold with Netra i." In his review of the Netra, Jaime Saker said, "Documentation is horribly poor." Michael's response is that, since the Netra was designed for a purely limited role, extensive documentation would have been a burden, not an aid. I might add that the documentation that came with my SS5 clone's SunOS media/documentation CD package would have been a big disappointment had I not been warned about it in advance; there is a single installation volume and nothing else. So you won't solve the documentation problem purely by getting a Sun system with SunOS. (It is claimed by Sun that the Solaris installation with the Sun Answerbook CD contains voluminous documentation; I have to say that I'd rather have some I can heft). Jaime Saker noted that the Netra has no C compiler included; this is true even of new Solaris systems. The Free Software Foundation has pre-compiled binaries of Gnu GCC for Solaris readily available. SunOS, incidentally, does include a C compiler that's good enough to compile GCC. Jaime Saker says that his early production Netra does not support video, even as an optional Sun board. Michael's response is that this is very logical, since the system is designed to be locked in a closet and not used. However, later Netras apparently no longer have this restriction; you can now attach video to it if you really want to. Jaime's comment on the "unacceptable" support is worth quoting in full, at least for its humour value: "Sun tech support, helpdesk and other support interfaces claim to have no knowledge of the Netra. (In fact, Sun's helpdesk was convinced that Netra must be a third-party software package. Describing the box with Sun's logo, the literature, serial/model numbers, etc. did not help. How can Sun support a product it denies exists?)" Michael concedes this basic point, but adds that support has now been substantially improved. Finally, Jaime makes this comment on his overall reaction to the system, as compared to other machines he evaluated: "Price-Performance ratio is dismal: Hewlett Packard HP 9000 712/60, equally equipped in terms of memory and hard disk, plus 15" monitor, a full OS, and a somewhat functional (enough to compile gcc/t++) compiler is LESS than the Netra with NO video, monitor, compiler, etc." Michael's response is that they put a great deal of effort into trying to make the Netra extremely easy to use, and that it is well worth the extra cost for Unixphobes and people who just don't have the time to learn Unix. So, although FAQ readers now know a lot more about the Netra than we did at first, it has to be said that the conclusion is obviously the same: No ISP worth its salt would buy a Netra; it's aimed at a completely different audience. For that audience, it's probably a very nice product. +.# Tony Sanders' Recommended Equipment List Tony Sanders, founder of the Inet-Access mailing list, has compiled some hardware information for use with BSDI systems. Much of this should also apply to Linux as well, although (to my knowledge), Linux does not yet have support for routing cards. I used to have an ancient copy of his document here. However, since it's available on the web that didn't make a frightful lot of sense, unless you really wanted to know what his list looked like three years ago. :-) Tony's latest document can be found at: http://www.bsdi.com/info/hardware-hints.etx +.# Other sources of information on PC hardware Eric Raymond's excellent PC-Clone Unix Hardware Buyer's Guide is available at http://www.ccil.org/~esr/clone-hw-guide/contents.html . Updates of Tony Sanders' listing of hardware for BSDI users can be found at the BSDI Web site, http://www.bsdi.com . Additional hardware and software information can be found in my Web site list for Internet providers, http://www.amazing.com/internet/ Bryan Taylor recommends Fintronics for pre-configured Linux systems. See http://www.fintronics.com/ . +.# What equipment is needed to hook up my system to the Internet? To try starting an ISP, all you really need is a 14.4kbps or 28.8kbps modem and a resellable SLIP connection to another provider. This is my current situation, as I try to gauge receptivity to my ideas and tune up the software. Unfortunately, resellable SLIP may be difficult and/or expensive to obtain. Aaron Nabil of internetworks (i.net) was kind enough to write a response with some interesting comments about resale. Actual resale policies vary depending on the vendor you select. "If you call one up and ask if they permit 'resale', to which they answer no, you shouldn't go away thinking they can't help you." Perhaps they can. Examples of different policies, from Aaron's message: * ANY TIME you make money from the connection, even if you just charge people to access a Gopher or WWW server, or sell products through that server. (Seattle providers) * Selling shell access to a computer connected to them is resale (Netcom). (He actually said "Netcom?", but I know from asking them that this is indeed the case). * Selling permanently addressed IP is "resale", transient IP isn't (Possibly Alternet) * Selling any IP is resale, but shell access isn't. (Internetworks [i.net, his company], most other providers * We don't care what you do. (Sprint, free-nets, ISI Network Associates). (I added ISI because I happen to know this is their policy). So, if you find a vendor and discover that they are not amenable to resale of their connection, make sure you know the precise definition thereof. You might be pleasantly surprised. You may also want to ask if they can "work something out"; special arrangements are possible. The next step above SLIP is a direct connection at a 56kbps data rate (double the speed of a 28.8 modem). The TLG Leased Line FAQ explains why a 56k is significantly better than a 28.8 SLIP; it has to do with latancy rates and other neat concepts I don't remember. I will later summarize portions of that FAQ here. TLG has some excellent FAQs and other information on connecting to the Internet at http://www.tlg.net/ . To deal with more than a very small number of customers on a paying basis, you will need a 56k or T1 connection hooked up to an internal network. For this, you need the following equipment: - A Router. This is a box that hooks up to your local network and sends out packets destined to the Internet, while leaving your local packets in your local network. You can theoretically program a PC (using BSDI software) as a router, but the rather vague impression I get is that this is for true experts only. Aaron Nabil of i.net comments: "Well, it's not necessarily for experts, but don't expect your carrier to help you fix it if it breaks." Tony Sanders adds: It's not really that hard and one advantage of using a system for the router is that something like a 486/66 or Pentium can pull double duty for a small provider with an eye on expenses and serve as a mail/www/gopher/ftp/news server at the same time as being the router (easily up to T1 speeds). Of course, as you move up in the world you might need to move to a dedicated router but I think that in general it's overkill for someone just getting started. For more information on BSDI hardware, see Tony's guest contribution, above. On the question of why Linux or other free Unix systems can't be used in the same way, Tony writes: "I think the problem is that they don't have support for any interface cards that talk to a CSU/DSU. Of course, that may have changed." Some inventive Linuxer is bound to change this eventually, but until then that seems to give BSDI a strong edge. Further information on the subject of using a BSDI box as a router is included at the end of this FAQ as Appendix A due to its extraordinary length (circa 400 lines). A counter-argument against using the BSDI box as a router comes from Scott Hinnrichs . There are two basic problems with this idea. "Routers should do just that, route. They shouldn't run Unix, or any other potentially lethal applications a hacker can get access through." Just as importantly, if you are tempted to use your BSDI box for anything else, you'll find yourself needing to take the system down for maintenance, which will kill off your Internet connection. A real router, on the other hand, should stay up forever, quietly humming in its pretty little box. He says, "I personally know of two BSDI/gated HDLC setups. One is still crashing/freezing randomly after 8 months (they've tried everything!). The other also runs PPP/SLIP access and other firewall support on the same machine, and it has frequent down times unrelated to routing." BSDI fans respond by noting that if you know what you're doing, it's reassuring to have the source code to tinker with if things go wrong, or if you need to add specialized features to your system. The best bottom-line remark I read about this came from Paul Vixie , who said that you should choose BSDI if you could figure out the source code and wanted to customize it, or a Cisco or similar box if you just wanted something that worked. In summary, the BSDI routing might be an acceptable solution if you have an ageing 386 around that you could dedicate just to the router, although even then security might be a legitimate problem. Certainly the router hardware and software shouldn't run on a machine that's used for any other activity. i.net is one of the few providers with the guts to list actual costs associated with the connection, including equipment. They charge $ 2,200 for a Cisco 2501 router. It can route packets up to the T1 level, and should do fine for most providers. "The Cisco is the benchmark of routers, the reference by which other routers are judged." Its main limitation is that it has only a single ethernet port and two serial ports, so it's pretty much limited to going from your internal network to the Internet. More expensive routers, like the Cisco 7000 series, can handle several such concurrent connections, and thus pass packets between several different internal networks. For almost all start-up providers, the 2501 should be sufficient. [I believe there are now routers that also serve as terminal servers; hopefully someone who actually knows something about hardware can contribute something on this vital subject]. For $ 1,700, Internetworks will also sell you a IRX-11, which will also handle 56k and T1 connections. "It's just not as nice a router as the Cisco." Avi Freedman recommends Bill Lunger as a source for used CISCO equipment and RAM. He can be reached at (612) 835-5502 and at bill@comstarinc.com. - A CSU/DSU. About $ 550 (again, using i.net's price sheet) for 56k or $ 1,300 for T1. This is the equivalent of a modem - it translates the router's output into a signal that can zip through the telco's lines. Aaron: There are cheaper brands for 56k, such as Adtran or Bat, that sell for about $ 250. "Try to get a 56/64k CSU/DSU if possible." Sean Shapira writes, "I have no complaints regarding my BAT 56k CSU/DSU." - A local area network to connect your router to your other computer(s). This could be an ethernet card in your system and Ethernet connections or 10Base-T, involving a hub and other stuff. (Someone else might want to flesh this out; this was modified through suggestions from Aaron Nabil and Sean Shapira). If you use your sole BSDI box as a router, you might be able to skip this, but most providers with a T1 or greater need a network to split their load between multiple machines. - Cables to hook everything together. They're easy to forget, but, according to Aaron, "they are non-trivial!" Depending on your specific Internet provider, you may need to provide the router and CSU/DSU at (a) your own site only; (b) at your site and their site. Some connections, called "full service", will provide the equipment and maintenance at both sites; a good example of this is CERFNet. This can be contrasted with ISI Network Associates, which requires you to provide the equipment on both sides. Naturally, ISI is a LOT cheaper than CERF. However, for a 56k connection, i.net is cheaper even than ISI, and they provide the router on their end. Go figure. +.# 56k Modems: Which way to go? There are currently two competing 56k modem standards, X2 and K56flex. X2 got out the door first, and is supported by US Robotics' popular modem lines. K56flex is supported by a consortium of the remaining modem makers. USR has used some particularly brutal marketing techniques to try and push ISPs into adopting X2. For example, they spammed entrants into their "Win an X2 modem contest" with information suggesting that they should ditch their current ISP and go with one supporting the X2 system. Because of this, most ISPs are extremely upset and not keen on doing business with USR. USR is the only company to support the X2 technology in their access servers (see below). The K56flex standard is both less expensive to the consumer and technically better, but modems supporting it are just coming out into the market. +.# Access Servers One of the most dramatic changes that occured since when I started my FAQ was the increased popularity of the access server, starting around the end of 1996. Access servers are integrated devices, combining modems and a terminal server in a single compact box. When I first started working on my FAQ, similar units were dramatically overpriced, and didn't actually do anything that couldn't be done by conventional modems. They were useful mainly for people with large numbers of phone lines, because they were a lot easier and cheaper to manage. So if you had a remote POP that was far away from your headquarters, and if a device that could save you from those tiresome trips and resultant service outages would be useful, then you needed one. Otherwise, you were better off sticking with a terminal server combined with regular modems. Two things have happened to change this significantly. * The prices of access servers have gone down dramatically. The early US Robotics Total Control racks cost around $ 40,000. Now, the universally loved Livingston PM-3 costs $ 16,500, for the same number of lines. This is not that much more expensive than a comparable terminal server + modems solution. * 56k modems are now a reality. Users, as you probably know from your own experience, are always hungry for speed. The new 56k modems will only deliver if they're talking to digital modems at the other end. So which access server should you buy? I talked to Justin Newton, who worked for Erols and bought thousands of lines worth, for his opinions. The bottom line is that the Livingston PM-3 is the best unit by far. It has been praised for its ability to work extremely well under load, and for the excellent technical support ISPs have obtained from Livingston. It's also considerably easier to configure than the rest of the competition, especially if you need to configure large numbers of units. The ISP/C (see elsewhere in the FAQ for information) has special discounts on Livingston equipment. In Justin's rather memorable words: "Livingston PM-3. Buy it, your customers, friends, relatives and children will thank you. Your pager will go off less in the middle of the night. They may even erect a statue for you in the town square. Analog modems are not recommended now that the 56k frenzy has started, and management of integrated digital dial access servers is much easier than the old "plug a modem into a serial port" days and ways. The PM-3 is the best of the bunch." The Ascend Max is the pioneering product, but Ascend users seem to be unhappy due to software bugs (initial releases are especially troublesome) and the inability of the product to perform under full load. The Cisco 5200 is said to be a fine unit, but again it has trouble running under full load. Support doesn't seem to be as friendly as Livingston's; you have to pay for a support contract in order to access much of the technical information). Now for the tough part. US Robotics, the leading modem maker and traditionally a company ISPs have appreciated for its overall product quality, makes a rack called the Total Control. Users of this rack have had highly mixed results with it. Some have said it's a fine unit, others have said it's inferior to the allegedly slower units it replaced in their shops. Many have found difficulties connecting with a large variety of modems, notably non-upgraded US Robotics sportsters. Worst of all, USR's dubious marketing tactics have antagonized people who really should be the company's friends. Overall, it's hard to recommend this product, even though it's the only rack to support the company's X2 technology. +.# What alternatives are there to access servers? Before the current age of the access server, you needed modems, telephone lines and one or more multi-port serial cards or terminal servers. Some of this information is now redundant; as I said, if you're serious about this business, you'll get access server systems like the ones described above. However, access servers have one major disadvantage: The phone company often charges extra for the incoming T1 connections. (These connections wrap together 24 voice lines per T1; they are not in any way related to the T1 you need to talk to your upstream provider. They're the same service from the phone company's point of view, but no, you can't use part of a T1 for dialup lines and another part for your upstream connection). In cases where you have a recalcitrent phone company with unreasonable charge for these T1s, you may wind up using modems anyway. In reality, you shouldn't, because modems are so tough to manage past a certain (small) initial number. But to start up, you may have to. No less than two people have written me with a simple question: What is a terminal server, anyway, and how do you connect all those lines to your PC? A terminal server is a device that has a whole bunch of serial ports and a single Ethernet output. The serial ports connect to your modems, and the Ethernet port connects to your network. Your network, in turn, is hooked up to the system that runs your user accounts. So people call in to the terminal server and connect to your machine. For reasons I still haven't quite figured out, this is far more efficient than a multi-port serial card. I think it has something to do with Ethernet protocols being more efficient than the direct handling of serial ports. It also has to do with the fact that the CPU is handling all the interrupt processing that is normally handled by a terminal server. There are newer, "smart" serial cards that can do a better job, but virtually all ISPs still recommend the separate terminal/access server. One major advantage of a terminal server is that you can program it to connect people telnetting in to one or more different machines. For example, if you had 3 systems for your shell or BBS accounts, you could tell the terminal server to route your first call to machine one, your second to machine two, and so on, thus equalizing the load between the machines. If one of them went down, you could instantly reprogram the terminal server to direct future calls to one of the other machines. This would be very difficult to manage with a multi- port card and modem setup. Walter Vose Jeffries has an interesting alternative, which has one major advantage: It's completely automatic. Use Call Forward if Busy and Call Forward On No Answer with your phone system. This way, if one of your lines is hung, the Call Forward on No Answer will send you over to the next line after a pre-determined amount of time. Curiously enough, the phone company has not been kind enough to tell me of the existance of this fine service; it may not be available in all areas, or it may apply only to business lines. It certainly sounds well worth a typical custom calling charge; it could reduce or eliminate the typical need for something like a USR Total Control style managed modem rack. The other way to hook up large numbers of modems to your PC is through a multi-port serial card. This is a simple card that fits in your PC. The servicing of this card takes a considerable amount of CPU time, although this can be somewhat relieved by buying an "Intelligent" serial card like the Cyclades. The big advantage of this approach is that it's a great deal cheaper than a terminal server. A 16-port terminal server costs around $ 2,000. A 16-port serial card costs around $ 250 (for the BocaBoard 2016; see my Boca FAQ for more information). A major disadvantage of this approach is that you're limited to about 16 ports per machine, 32 if you're lucky. A Sun SparcSTATION 20 using terminal servers can support around 150 users. So if you factor in the cost of extra computers per 16 or 32 lines, the terminal server starts to look like a more reasonably priced solution. MODEMS: Many people swear by rack-mount USR or other big name modems. You can do all sorts of neat things with them, such as remote test and reset of individual lines. Unfortunately, they cost about triple what low-end standalone units do. Low-end standalone units are more likely to not connect properly after a disconnection. This appears to be especially true of US Robotics Sportster units, although their high-end modems are superb. I've had excellent success with Intel 144e modems, which cost all of $ 99 each. One note on the Intels: They have apparently changed their design significantly in the last few months. The newer ones are instantly recognizable by their roughly square power bricks; the older ones were clearly rectangular. The difference relevent to Internet providers is that the older ones support "at&q6" to turn on error correction and autobauding. The comparable command for the new series is "atb0\j1". The \j1 turns on autobauding and the b0 specifies CCITT mode. Note that the "\j1" will have to be typed in as "\\j1" on most Unix systems. Unfortunately, Intel is apparently getting out of the modem business. I, for one, will miss them. Alicia Salomon (salomon@seas.gwu.edu) was kind enough to send me a price list for US Robotics rackmount modems. I have to say they seemed pretty forbidding, even with special "Internet Pricing". The basic chassis, including "dual power units and the Network Management Card Set (NIC/NAC)" is $ 3,810. From what I gather, this gives you the ability to reset and reassign modems remotely by connecting to the modem's ethernet slot and giving them commands. You then must buy a Dual T1 card set for $ 2,701 to hook these modems to a T1 line, which can then be brought in to your network. A modem card, which contains four modems, costs around $ 2,000 (again with the special pricing). So if you wanted to start with 16 lines, it would cost a eye-popping $ 14,511; 16 Intel 144e modems would run only $ 2,240. Filling the box to its 48-line capacity would cost $ 38,511, while 48 144es would cost $ 6,720. However, this is not quite a fair comparison, since this price apparently includes the equivalant of a terminal server, which would otherwise cost somewhere around $ 2,000 for 16 lines. Despite this, I suspect the rackmounts have their place. If you have a POP that's remote from your main business location, you might not be able to go there and physically reset the modems in any reasonable period of time. With that situation, rackmounts might actually be the best solution, since you'd just reset them via your network. An alternative would be to build your own remote switching device, so you could remotely switch the modems on and off when they needed to be reset. Even hiring someone to design and build such a thing might be cheaper than a rackmount modem box. People on the Inet-Access mailing list have recently said some very nice things about Ascend products, in particular the Ascend Max 4000. This device can handle up to 48 phone lines and/or 96 ISDN BRI connections. Apparently providers who use this solution swear by it, and it certainly couldn't be more convenient; it wraps up modems and terminal servers in one box that's about the size of a good stereo power amplifier. One very neat thing about the unit is that it can handle any kind of incoming call. The way it works is this: You get ISDN-capable lines from the phone company, and they give you a single telephone number. Both your analog and ISDN customers use this phone line and number to dial in; the Ascend box simulates an analogue modem when connected to a modem and an ISDN line when connected to an ISDN modem. So you can use the same set of phone lines for everything. Unfortunately, this is again a fairly expensive solution, but a key advantage is that it interoperates smoothly with all other modems on the market, a key advantage. Most lower-end external modems will operate best with a relatively narrow range of manufacturers. Recently, there have been significant complaints about Ascend service and support on the Inet-Access mailing list, so I would go with the Cisco products (below) instead. The Cisco 5100 and Cisco 5200 are similar products, known for superb quality, excellent support and mind-bending cost. "Empty your wallet and your trouble log" says one Inet-Access subscriber about these units, which can support 48 lines for about $ 37,500. I have been told, however, that the 5200 can be seen for around $ 20,000 (instead of that $ 37,500 list price). The 5100 is apparently based on USR Total Control technology; the 5200 is a combination of Cisco's own designs and Microcom modems. The difference between the two units is that the 5100 contains more or less conventional rackmount modems. The 5200 includes a dual T1 card that receives two special phone company T1 lines that contain 24 multiplexed connections each. So you have two connections in to hook up to 48 phone lines. (Note that this cannot be confused with the T1 line that you use to connect to your upstream provider; the phone company products are essentially the same, but their uses are totally different, and they cannot be combined). Sean Shapira notes the primary advantages of rackmounts: "space savings; significantly reduce wiring harness". Bryant Durrell has some interesting advice for the startup ISP with multiple POPs: "Something I'd be exploring if I were a small ISP opening my first new POP is colocates. There are a lot of people out there who'd love to have a POP in their house, just cause it seems cool. Some of them may be technically inclined. I don't know if one could find trustworthy people, but it seems possible." Eric S Raymond took the opposite approach in his system setup, which I think will appeal a great deal more to the startup ISP. His contribution also gives some us some idea of the kind of patching BSDI users have had to do. He writes the following paragraph about his experiences: "We got our nonprofit ISP started using a super-cheap modem called a LineLink 144e, built around the Rockwell data-pump chip (same one used in the Zoom and Boca modems) and costing $105. We hooked our modems to SDL RISCOM/8 multiport boards, the brand recommended by BSDI. This setup has worked pretty well, except for one major problem -- when UNIX on our 50MHz box hangs up, the DTR-low interval goes by so fast that the LineLink sometimes fails to see it. This causes the modem to hang in the off-hook state, blocking the line and requiring a manual reset. I worked around this by patching a 250msec delay into the RISCOM driver's DTR-pulldown code. This fix may become unnecessary when RISCOM releases the next driver version, which is supposed to do true hardware handshaking on the modem lines." Incidentally, the setup I had under Linux, featuring a BocaBoard 2016 16-port serial card and $ 139 [now $ 99] Intel 144e modems, has worked flawlessly with no installation problems at all. MULTI-PORT EQUIPMENT: Your modems have to connect to your computer, which normally has two or fewer available serial cards. How to do this? There are three basic ways: DUMB CARDS: These cards give all processing to your system's CPU, which makes it run slower than the alternatives. However, they are very cheap and relatively easy to set up. If you have a Linux system, you should request my BOCA-FAQ, which outlines the procedure for setting up a Boca 16-port board under Linux. SMART CARDS: These cards take some of the processing load off the main system. They are supported by the BSD systems, but not Linux; as a result I have limited knowledge of them. TERMINAL SERVERS: These are high-end products that are mercilessly expensive, often over $ 2,000. In return, you get a device that handles your terminal ports by effectively telnetting to your system. This is significantly easier on your system, since it no longer needs to process any form of terminal interrupts. This is, however, yet another item I could use some help on, since I've never used one of these beasts - too much money! ACCESS SERVERS: See the comments on the Cisco 5100/5200 and Ascend Max, above. Karl Denninger writes the following on various brands of terminal servers. (This was in 1994, so this is pretty obsolete). Annex: Unix-style kernel. Well-known, venerable, highly stable. Decent performance, but you will NOT be able to drive all the Annex ports to full speed at once. I've tried it. Livingston: Newer, good reputation, but has a few problems that I can't live with (primarily no host route advertisement). They claim to be fixing this. RADIUS authentication system is quite nice. Not a bad box. Telebit: Venerable, well-understood, *extremely* flexible command set and capabilities. Can handle leased connections as well (up to T1 with appropriate cards) which makes it a "POP in a box" possibility. Classics are available cheaply, current units (NB40s, etc) are more expensive. Will route IPX and Appletalk in addition to IP. Basically it's a question of what you want and need. +.# How many phone lines do I need? To start, it pretty much depends on your budget. I currently have four: three incoming lines plus my SLIP connection. Because my system is experimental, nobody calls it yet and so I have no hard answer to this question. (After a few months of running software that works reasonably well (not fully debugged by any means), my lines are now often full). My estimate is that you want 8-10 lines to start, once you're ready to give your system a bit of publicity. But it really all depends on your market and how high a profile you can maintain. Since I've written this, I've started getting occasional busy signals on my three-line system. I have about 100 user accounts and 10 people who call several times a day. Since I don't charge for the system yet, however, most of these numbers are meaningless. Stay tuned. As a general rule, 10 users per line is suggested for conventional dial-up connections. I believe Karl Denninger maintains roughly this same ratio, even with his SLIP connections; he can do this due to a 20-minute idle timeout for the SLIP. Alicia Salomon adds, "after about 400 users, it goes to about 12:1 and then goes to 15:1 around 1000ish (guesstimates based on vague sources of data input)". If you have under 32 lines on you system, you may wind up having to buy a line for every 6-8 users. Permanent SLIP connections by definition take precisely one dial-up line per user, and should be priced accordingly. Some people have gone to 4-6 users per line even for non-permanent SLIP. George Herbert has a good summary of what happens when your modems go over that magic ratio: "'Good' services will have a ratio of 10 to 12 users per modem. At this level, you generally will not see busy signals except for brief periods of time during peak hours (which are usually 5pm-midnight local time). Users seem not to mind at all if they get a busy signal for a couple of minutes every few days, so it seems to be OK. "At a ratio around 15 to 1, you see people talking about longer periods of busies (10+ minutes) regularly every night, and you start to get complaints. "At 18:1, your users start defecting en masse as they can't get on for hours on end ... the worst possible example was 20:1, briefly, for a major service who I won't name, which led to the other major services in town picking up several hundred defecting customers and loads of public postings of displeasure." Steve Balbach reminds us that lines can take a long time to install. "We give 2-4 months lead time from when we decide to add more lines to when they are live. Some examples of time delaying problems: v.34 chip shortages industry wide put new modem orders on hold Bell runs into facility problems at your location. Bell messes up your order and takes weeks to straighten it out. Electrical upgrades required Wiring upgrades UPS/power backup upgrades I'm sure there's a slew of other possible problems that can arise. If you're at 12:1 now and decide to put new modems in, you're too late - expect possibly a few months of busy signals. And add more lines than you need , proactive is the key." I suspect this is especially good advice for a large ISP that runs sizable numbers of lines and has to order modems in bulk. I've certainly never had any trouble getting my modems from Fry's electronics or some similar vendor. However, I once had a major crisis when the phone company ran out of lines to put in my home; it's taking about two months to straighten that out (through my upcoming move to a commercial location). However, if you use certain items of popular hardware like US Robotics Total Control modem racks, you have to contend with the potential for major shortages. *.# What about Residential phone lines versus business? Residential phone lines are generally much cheaper than business lines. Traditionally, the phone company has subsidized the cost of residence lines by charging businesses substantially more for phone service. Fortunately, the network provider has almost exclusively incoming calls, so the measured service issue for business lines doesn't come up much. In some areas, the base rates for business lines are about double residence rates. In other areas, they are about the same. In most places, all calls are measured, so you should watch out when calling from any business lines. Whether you use residence lines or not obviously depends at least in part on the location of your business; you cannot use residential lines if you're in a business location such as an office building or store. Phone company policies on use of residence lines for a business operating out of people's homes vary dramatically depending on the area in which you live. The worst case is that you could be back-billed for business rates from the inception of your business. Best case, of course, is that the phone company could ignore you completely and allow you the residence lines without trouble. Fortunately for the budding ISP, phone companies in many cities are setting up telecommuting programmes and in general trying to encourage people to work at home. One of the major elements of this appears to be allowing home businesses to "get away" with the use of residence lines. Because of this, the pressure to get business lines seems to have abated in many areas. For example, a few years back, Pacific Bell tried to institute a crackdown against people who were using residential lines for their BBSs. However they have recently reversed this policy; now, they are more than happy to take orders for multi-line BBSs at residential rates; my rep knew exactly what I was doing when I placed the order, and even offered a few words of encouragement. When I asked if there might be trouble in the future, my rep - who seemed like quite a knowledgeable fellow - said no. According to Alan Byrant's book on running a successful BBS, Southwestern Bell has been particulary aggressive at nabbing providers and BBSs who try running business systems off residential lines. (I read this in the bookstore and unfortunately don't remember the name of the book). Unfortunately, there is a serious snag that I came across only when I ordered a new 56k connection from Pacific Bell. In Pac Bell territory, if you go over six lines, you are considered responsible for any wiring costs to get additional lines out to your house. So in order to get a seventh line, I would have to pay the $ 9,000 charge to get additional lines wired in from the central office. For a single seventh line, the charge might be lower, but since I needed a large number of lines (about six more then), there was no really cost-effective solution other than going to a business location. Unfortunately, the phone company is less than forthcoming when asked about the number of lines that can be put into a house or apartment. As a general rule, they will actually attempt to do the installation, find out they don't have enough lines available, and then you're stuck. One possible option is called a "Mux" or "Channel Bank". The phone company gives you a T1 line between you and the central office; it can hold up to 24 voice connections. You can then use the channel bank to split the T1 into the 24 lines at your location. The problem with this solution is that the channel bank itself costs about $ 4,000, and it will only give you 24 lines per two that you presently have. So if you can have a maximum of six lines, and two are in use already as voice lines, your channel bank could give you 24 lines, and your other two lines would be used by your Internet connection (56k or T1). So your maximum growth at that site would be 24 lines. Note that if you get a Livingston PM3, Cisco 5200 or Ascend Max, you are automatically using this solution and you need to get the special lines from the phone company. According to Joe McGuckin , a company called Carrier Access Corp makes a $ 2,995 channel bank that (unlike the previous generation) fits into a standard rack and runs on 120VAC (or -48VDC for rack fans). It also comes in a 12-line version for $ 2,195. Definitely worth checking out, according to Joe. (Note that you probably wouldn't want to put those final two voice lines in the channel bank, because they're residential voice lines, not business lines; you want them to be free of toll charges). Pacific Bell does NOT charge extra for this arrangement; other phone companies do. When you're thinking of going this route, make sure you know ALL possible phone company charges; otherwise, they might bite you in unexpected places! +.# What's this I hear about cut-rate 800 line service? In a word, it's a scam. According to Jeff Woods (jeff@delta.com): I got taken by this one a few months back. IT IS A SCAM. When they "have to change numbers" it is because of their illegal activities. These suppliers "buying unused LD time" (there is no such thing) are in effect taking your money, and getting a REGULAR AT&T or MCI number for you IN YOUR NAME, but billed to a bogus address. Three months later, when YOU have not paid your bill, that company cuts the service off, and the "supplier" simply gets another one for you. We lost $2500 to a company doing this in the spring. The perpetrator, one Greg Evans, had sucked in Global Networks of California and a group of bankrolling lawyers in Ann Arbor, MI (who are also the ones selling support for $1 per user). When the feds closed in on Evans, he skipped town, leaving behind MILLIONS in unpaid 800 bills, and having taken the ISP's for their $5000 signup fees and $6 per user per month. He skipped town (they never caught him) with hundreds of thousands, and apparently just set up again elsewhere, with a few dozen copycats. Notably, though, the ISP's were the ones stuck with these monster phon bills -- AT&T claims that the numbers were in the name of the ISP, not of the "supplier", and as such, the ISP was responsible. Do *NOT* fall for it, people. You WILL lose money, and could go to jail for fraud. AT&T and MCI are actively investigating any instances of this. If you suspect that a competitor is caught up in this, and KNOW the 800 number they're using, forward it to a long distance company's security department. They will take it from there, and will see that it reaches the proper company (they cooperate with each other to keep fraud losses low). YOU HAVE BEEN WARNED. THIS IS A SCAM OF MONSTER PROPORTIONS. See also http://www.techweb.com/wire/news/jul/0729att.html . +.# Where do I put all this stuff? Most beginning ISPs start in their owner's home. This is nothing to be ashamed of; even mighty Netcom started this way. Of course I'm just a shade biased here; my nascent provider is right here at home, too. The home address has some very interesting advantages: (1) It doesn't cost any more than what you're already paying for rent or mortgage. (2) You can use residential phone lines instead of business; rates are 50% or less business costs. However, see above for a full discussion of the issues behind this. (3) It's easy to get to in case of an emergency. Just walk to your desk. It also has some disadvantages you might not be aware of: (1) It's probably technically illegal, thanks to zoning rules. Fortunately, these are being gradually loosened. (2) The tax consequences of deducting the portion of your mortgage used for business purposes are extraordinarily murky - and whether you rent or buy your home, office in home deductions are a major red flag for IRS audits. (3) It has a few image problems. You don't normally want to escort potential customers for big accounts into your living room and have them run into that strange mess the kids created during playtime. Potential employees may also be less than impressed by your working environment. (4) Some telco services may not be available to residential customers. This is one issue I don't know much about - I'd appreciate more information from people who know what they're talking about here. (5) You may be able to bring in only a very limited number of lines; see the previous section. In the end, though, it all boils down to money. If you have tons of the stuff, you'll probably have an office. If money's tight, running your ISP out of your home is one of the best ways to save. +.# Choosing your location: Disaster planning and Phone Line Questions There has been an interesting recent thread on the inet-access mailing list about what location is least prone to natural disasters and other mishaps. These could range from a mighty earthquake to a raging fire. The ideal solution would seem to be a building you own yourself that's not shared with any other tenants. The non-shared aspect would mean that you wouldn't be vunerable to man-made disasters caused by them; the building would probably be only a single story and thus less vunerable to earthquakes. Ideally, the building should be separated from others (like a single family house as opposed to a building in a bunch of cheek-by-jowl commercial buildings). After the recent Oklahoma bombing, it's pretty clear that you don't want to share your building with government offices. Unfortunately, few of us can afford such a costly solution, unless we ran the business out of our homes. People who run an ISP out of their homes may find problems installing T1 lines (which may not be cost effective to the phone company) and installing large numbers of phone lines. Most commercial buildings have no set limit to the number of lines that can be installed, although it's always a good idea to ask the phone company before you finally select your building. There would appear to be an enormous variance in the number of phone lines you can install in a residential building, ranging from 3 to 300. The upshot is that good relations with the phone company are essential - however much you may swear at them behind their back. See directly below for a major advantage of starting your service in just the place you might not expect to be cost-effective: A glossy, expensive high-rise building in the most prestigeous area of town! +.# What about competition in local phone service? Once you've recovered from the shock of the cost of a T1 line, your next major expense is the cost of local phone service. Depending on your area, the cost of, say, 200 lines can range from $ 4,000 a month to $ 7,500 a month. That's quite a arange, and that's just between Pacific Bell (the low figure) and General Telephone (the high). So, you might wonder, what about the competition we're supposed to be getting as part of the telecom reform bill? There has to be better news than the CDA, no? The good news is that a company called MFS - http://www.mfs.net/ - can provide you with local phone service. When I talked to a MFS rep on the phone, he was positively glowing with a desire to be helpful. The main catch with MFS is that they are primarily offering access in MFS-wired buildings. So when you are considering a location, you might want to make finding such a building a priority. They are normally large, high-rise structures located in prime areas of major urban areas. Note that being in a MFS building can save you in two ways: You'll get cheaper local loop charges for your Internet connection, as well as less expensive local phone service. In return, of course, you'll pay higher rents for significantly glossier office space than you might expect - but unless glossy office space makes you feel acutely uncomfortable, just think of all the money you'll save! I haven't gotten all the information I need yet from MFS, but so far I've been impressed by their friendly and cooperative attitude. +.# What can I do if my customers get a lot of line noise? The traditional answer is, "Not much." The phone company is typically pretty unsympathetic to complaints about substandard connection quality; the typical response is that, if you can hear a clear voice signal through it, it's fine. Jack O'Neill [jacko@onyx.xtalwind.net] found out something quite fascinating, however: If you claim it's your FAX line and you'll drop it if you don't get satisfactory results, they will clean it up. A digital phone line, such as the ones that use channel banks and/or Cisco 5200/Ascend Max style units, is significantly cleaner than analogue lines. +.# This is so expensive! Where can I find this stuff cheap? (*) Don't forget the newsgroups misc.forsale.computers.workstation and the corresponding pc-clone groups for the best prices on hardware. Unfortunately, much of what's offered is rather low end and not really suitable for an ISP. I did get my Sun 3/60 through the workstation group, but it's not going to power a full provider; I use my Linux PC for that and use the Sun for its neat 19" colour monitor. There is a mailing list for used equipment called ISP-Services. To join, send an email to listserv@ispc.org with subscribe isp-services in the message body. More recently, some very interesting hardware (SS10s) has been offered with some regularity on the workstation group, so watch for them and pounce if you want a SS10. If you're looking for PC-style hardware to run your provider, Computer Shopper is one of the best sources of deals. If you're interested in Silicon Graphics, my personal favourite platform, be sure to wander over to Reputable Systems at http://www.reputable.com . Greg Douglas there has been extremely helpful to me. You may also want to check out my SGI buyers' guide at http://www.amazing.com/internet/old-sgi-faq.html . For Sun workstations, an excellent place to start is by dropping a line to Craig.Warner@ceram.com. He's a Sun clone dealer, and my experience with him was excellent, even though I didn't wind up buying anything (yet!). For real Suns, you'll have to find a local reseller. The one I talked to could not relate to my needs, but hopefully you'll have better luck. Unfortunately, Craig no longer works for Ceram and I keep on losing his email address :-(. His successor at Ceram is Gary Patterson (garyp@ceram.com). Henry Minsky (hqm@ai.mit.edu) is developing a list of dealers who carry used equipment. Access it through a Web browser via the URL http://www.ai.mit.edu/datawave/hardware.html . "There are some pieces of equipment I would not recommend getting used, but others, such as the tape drive, have warranties from the reseller, and seem to be a good option." David K Merriman suggests the following dealers: Personal Computing Tools | Data acquisition, good source for 90 Industrial Park Road | Digiboard multi-port cards Hingham, MA 02043 (800) 767-6728; fax (617) 740-2728; BBS (617) 740-0061 Data Comm Warehouse | Network cards, hubs, routers, patch panels, 1720 Oak Street | RAS hardware, modems, cable, connectors, LAN P O Box 301 | software, LAN test equipment, cabling tools, Lakewood, NJ 08701-9885 | UPSs, patch panels, rackmount, etc (800) 328-2261; FAX (908) 363-4823 Comment: "Have dealt with them, and they are very good on delivery and support. First catalog I turn to for network stuff." + Hooking up to the Internet +.# How do I hook up to the Internet? By finding an Internet provider in your area who's willing to hook you up, or by connecting with the major services such as Net Access, SprintNet or AlterNet. +.# How does the Internet work? Every once in a while, someone drops me a line in my mailbox asking a question like, "Why do I have to pay the phone company to hook me up and a wholesale provider to connect me? Why not hook me up directly to the Internet?" Because there is no such thing. The Internet is not a single person or body, but rather a whole bunch of people and companies working together. The Internet is composed of complex interconnections between several different large companies, who do the long distance routing for you. If you're a typical person reading this FAQ, the odds that you could participate in this high-level long distance routing are practically nil. You would need a Cisco 7500-series router, which costs over $ 100,000, as well as highly trained engineering help. Effectively, you would have to compete with Sprint or MCI, with your own long distance lines and such. A company called Net-99 was able to start up by leasing lines from a company called Metropolitan Fiber Systems (MFS). They had a fairly prosperous run and then they got taken over and absorbed into the AGIS network. After that, most of the original partners either left or were booted out. If you're determined to start your own backbone provider, though, history appears to show that it can be done, if you have deep enough pockets. However, if you're considering operation of an ISP, be forewarned that being a backbone provider is far more complex and challenging. My advice to you would be to try to walk before you run. All these long distance companies peer - that is, they hook up to each other - at places called Interchange Points, such as MAE-East. This is why the traceroute command often shows very confusing routes from one place to another. For a while, MAE-East was the only reliable way to go from one backbone (long distance) provider to another, so a packet travelling from Los Angeles to the San Francisco bay area would go all the way across the country on one provider, whiz through MAE-East, and then go back across the country again to San Francisco. As an individual provider, you have basically no chance of being able to successfully conclude peering agreements with the big boys at one of these sites. And if they don't have a concrete reason to talk to you - either sites their subscribers want to connect to or payments from you to them - they won't. And bear in mind that if you're at a major interchange point, you will need to peer with all the carriers there to get full coverage - not just one. +.# What are the levels of connection you can buy? ------------------------------------------------------ Maximum Connection Equipment Required Approx Cost Simutaneous Users 28.8K SLIP 28.8k modem $ 300 around 3 56k Leased CSU/DSU 250 around 8 Router -or- routing card 2300 or 500 T1 Leased CSU/DSU 1400 around 200 Router -or- routing card 2300 or 500 ------------------------------------------------------ Some connections will require that you buy a modem, router and CSU/DSU for both ends of the connection. ISI Network Associates, for example, requires this, meaning that their 56k connection would cost their $ 1,000 startup fee plus two times $ 2,550 - a total of $ 6,100. As an alternative, you could pay a $ 4,000 startup fee and still buy the equipment on your end. Other connections furnish the equipment on both ends, including a service contract. Unfortunately, the latter type of connection is normally prohibitively expensive. The telephone company rates for these connections range from reasonable to horrendous. On the low end, the phone line I use for my 28.8 SLIP is a conventional residential line which costs around $ 20/month. Leased line fees for a 56k line range (in Southern California, anyway) from about $ 100-200/month. T1 fees range from $ 400 - $ 1,200 a month, depending on the distance between your site and the nearest provider POP. +.# What is Frame Relay? The following Information about Frame Relay connections was contributed by Sean Shaprio ; unfortunately my explanation is sufficiently changed from his original words that errors are particularly likely. He suggests that you read "the little 30 page book (published by Motorola?) that describes it in excellent detail" if you are seriously considering this. Frame Relay connections are an up and coming form that are closer in nature to a switched telephone connection than a traditional 56k/T1 link. To connect to a provider with frame relay, you run a leased line to the nearest frame relay access point. The connection is then made to your provider, even if it is a very long distance away. The provider runs a high speed connection to his nearest frame relay node, where it can then get split off to several connections. So instead of having multiple 56k connections to his customers, he pulls a single T1 into a frame relay switch; the packets are then switched over to the customer's 56k hookups. This is the service that lets providers like Netcom and CRL operate nationally while still having all their equipment in their original Bay Area headquarters. Dave Van Allen was kind enough to contribute a more comprehensive explanation of Frame Relay than we had in the past. Frame Relay's main advantage is that it's a cut-rate form of leased line service. "For instance, in the Northeast Bell Atlantic region, 56k service has an initial install of about $ 200 and a monthly fixed cost of apx. $ 150. You Tools [his provider] offers 56k (with a 56k CIR) access for $ 399.00 per month, including the Internet access (Port) and telco charges. The customer premise equipment (CPE) [the equipment the customer has to buy to use the connection] is about the same for Frame Relay as it is for ISDN - approximately $ 900." The main problem with frame relay is that you are sharing a switched line with a potentially large number of other users; this line has a fixed capacity that is divided between you and them. Your provider will give you a committed information rate (CIR), which tells you the minimum speed your line will connect at, no matter what the conditions. Sprint presently gives a CIR of zero, which means that they do not guarantee that your connection will be continuously running at any particular speed, or even operational at all. So in theory, if you were running a Frame Relay T1, you might have a virtually worthless connection if all the other connection users took up all the bandwidth. There may be something seriously wrong with the above paragraph. Dave Van Allen again: "In the FAQ, you reference Sprint's CIR of '0', meaning that they don't guarantee any performance. This is not the case as I know it. The CIR of '0' indicated that the FR link has NO committed information rate, and the link is specified to run at the bandwidth sold. So, a 56K FR link, with a CIR of '0' (from Sprint, at least) will run at 56k baud at all times." The reason I haven't lifted the previous explanation from the FAQ is that it seems to fit the facts; Karl Denninger has been complaining about the "0" Sprint CIR for some time. Certainly the complaints I have read about Sprint's service seem to imply that the "0" CIR may indicate a lower quality service commitment. However, from a technical standpoint, Dave may well be right. Dave continues: "With Frame Relay, the FR provider has a system of switching ports that share the bandwidth of the Frame network. Because a switched packet network is a non-dedicated data path, the equipment used to provide the service is normally the limiting factor in just how much bandwidth 'everyone' can have. If the provider is under-utilizing the capacity of the switch, then ALL traffic in that switch may always travel at the highest speed -your- line can handle. "If traffic in the switch gets heavy, then the provider must either add more bandwidth, or limit the speed of the connections during peak periods. This is what the CIR is. CIR is a provider imposed limit on the speed of your connection. Most phone companies sell like this: You get a 56k FR connection with a CIR of 30k - this means that you will be guaranteed that the connection will go at least 30k and will peak at 56k. In reality, in most areas, that connection will do 56k and might rarely drop to 30k for brief periods." "If you purchase a 56k FR connection with a 30K CIR, you can often request that they special-build you a 56k CIR connection. There will be a small surcharge for this, but it is possible. "Frame relay can work up to T-1 rates, and it is usually the least costly option at those speeds." The advantage of Frame Relay should be fairly obvious; since you're sharing a large connection with other people, you aren't paying the phone company for hideously expensive leased lines. Because of this, Frame Relay is a much cheaper service than the traditional 56k or T1 leased line. This form of connection requires a special frame relay compatible router to work. * How is the performance compared to standard 56k/T1? Karl Denninger and his partner Joseph Stroup initially decided against offering Frame Relay with their Net-99 service due to low connection quality. However, they are currently offering a pilot Frame Relay project at an attractive rate; they will expand it if it meets their quality standards. One other major advantage of Frame Relay is that the phone company offers attractive rates for different levels of service that can scale up as you need more bandwidth. For instance, in Pacific Bell territory, Frame Relay costs $ 333 a month for 128 and $ 650 for full-bore T1 with a couple of levels put in. If you start with 128k and max out your bandwidth, you can convert to one of the higher levels for a $ 30 charge and, of course, payment of the higher monthly rates. An update on this section comes from Roger Books (books@mail.state.fl.us). He writes: "CIR, Committed Information Rate, the bandwidth gauranteed you by your provider, burst is allowing you to use bandwidth above this, if your CIR is 0 then everything is burst. Do note that if you are using frame over a dedicated link the standard is to leave the CIR to 0 on the assumption that if nobody is competing for bandwidth there is no reason to setup CIR." That seems to clarify the issue. +.# Who are the main national providers, and how much? This section used to have quite a few more prices than it does, but now you can determine prices quite easily by searching the web. This is especially important since they're changing all the time. What's right today will no doubt be wrong a week from now. Net Access - http://www.netaxs.com/ is a new backbone providers being run by people with a proven track record in this industry. Net Access is already in operation. Net Access has a long history of providing bandwidth around the Philadelphia, PA and Washington, DC areas. They are now expanding to nationwide service. Net Access is also Philadelphia's oldest ISP. Avi Freedman, owner, is best known for offering some of the highest quality free advice on Inet-Access, and for providing the list with some of the best much-needed comic relief. This FAQ is largely sponsored by Net Access. I previously mentioned priori.net as a promising business venture. Unfortunately, the company is now approaching Chapter 7 liquidation, meaning that it will cease operations shortly, leaving a great many people out of connectivity. I'll have some more news on this when I find someone willing to talk about it. :-) ANS has been around for years and years, but for most of that time their pricing was hideously uncompetitive. They made most of their sales to corporate end-users, who need a higher level of reliability than the average, and are willing to pay for it. Recently - around June of 1997 - they have brought forth a push into the ISP biz. They now charge about $ 2,300 a month for a T1, which is highly competitive with other national providers. Avi Freedman of Net Access, another company going into the national backbone market, says: ANS is, in my opinion, the highest-quality ISP in the known universe. ;If they sell T1s for under <$2500/mo they should be seriously considered. Like, an order of magnitude better than any other national provider out there. MCI has made a really big push into the Internet market. Of all the major providers, it now seems to have the best reputation for technical quality. In fact, you will notice that it's doing a healthy percentage of the net's long distance communications now. They are legendary for having the industry's worst billing practices; expect hopelessly garbled invoices. They also have at times been accused of trying to begin volume charges for their bandwidth; what that means in this competitive market is anyone's guess. Rumour has it that they have started publishing such charges, but their precise nature has yet to be determined. Kevin Kadow says that AlterNet will sell to resellers at $ 750 over their quoted prices; my previous FAQ draft said that they were not selling to resellers at all. However, they are "getting more buddy-buddy with Microsoft after their deal." However, a couple of people have defended AlterNET, saying that they have the best service quality of any Internet provider. In mid-1997, this was largely contradicted by some sweepingly extensive service outages, some of several hours' duration. Note that "AlterNET" and "UUNET" are the same company. I recently commented to the Inet-Access list that Sprint and other large net providers have a reputation for poor service. I then heard from three different Sprint users who professed complete happiness with the service, so it looks like they have cleaned up their act. Only a few months ago, I heard bad things about them at about the same rate. One observer says that MCI and Sprint's reputation seems to vary significantly by service city, so you may want to check out their reputations in your area before signing up. If the national provider rates seem too expensive, you may wish to hook up with a local provider. Watch out for the connection quality, though; if the local company sells you a T1, and all they have is a T1 connection themselves, you're bound to get mightily poor throughput to the rest of the Internet. You might be especially happy dealing with a local provider service if it has multiple T1s to more than one net provider. This will ensure that you don't lose connectivity even if one of their downstream providers has trouble. If you happen to be in the San Francisco Bay Area, The Little Garden (TLG) has an excellent reputation and very fair rates. For that reason, the Bay Area is largely overloaded with Internet providers, so unless you have a very special business plan, you may not want to set up there. For information on Net-99/AGIS, see the next section. +.# What happened to Net-99/AGIS? In late 1994, Net-99 was started with a tiny amount of money (reputed to be around $ 30,000), some shadowy figures in top management, and the entire personal and reputational capital of one Karl Denninger, a legendary success story. Karl Denninger had won considerable points here by starting a new backbone provider at a time when backbone providers were all overburdened, overwhelmed and generally poorly managed. Karl's Macro Computer Solutions (MCS) had always been an impressive operation, and his involvement with Net-99 was said to be the guarantee that it would provide service with integrity and good will. Naturally, Karl didn't disabuse anyone of this impression, and neither did anyone else in Net-99. Behind the scenes, though, Joseph Stroup, a considerably less impressive figure with quite a bit less in the way of networking credentials, was really running the show. (I still remember the striking contrast between Joseph Stroup's illiterate three line messages and Karl Denninger's 100 line plus masterpieces of the "drive the point home, and then drive it in again and again and again" school of argument). Net-99's technical staff was of high quality, but the network itself was being held together by Cisco 2501s, which were not ready for the task of handling a major national backbone. Despite this, customers were generally impressed with the service. One major industry figure told me that it was "like a cult." In summary, people were so relieved not to be dealing with incompetent Sprint, questionable MCI or overpriced UUNET that they'd take pretty much anything as good service. lDespite this apparent success, a little under a year into the history of the organization, constant mismanagement at the top caused the situation of Net-99 to decay. Joseph Stroup sold the company for approximately $ 1,500,000 - an absurdly low amount for a company in a rapidly growing business with nearly $ 2 million in sales, but quite a nice return on the $ 30,000. Karl Denninger, incidentally, got nothing and was, to say the least, rather upset. This was especially true since the most powerful routers used by Net-99 were actually on loan from him, and because his image as loaned to the venture was one of the things that drove the venture. But no written agreement, so no share in the proceeds. So AGIS took over Net-99, and almost immediately realized that they'd bought an inadequately managed and equipped service, in striking contrast to its public image. Unfortuantely, they didn't recognize the quality of the Net-99 staff, and virtually all the top technical people left the company for greener pastures within a week or two of the takeover. This left AGIS with a network but nobody good to run it. Things got so bad that a mailing list of dissatisfied AGIS users was created. with what I've seen classified as highly contentious discussions. AGIS rapidly went from one of the best places to buy bandwith to one of the worst. The next episode in the history of AGIS stars the legendary Sanford Wallace, owner of Cyber Promotions, the spamming firm. If you've received mail from savetrees.com or cyberpromo.com or any number of other domains, you've gotten spammed by Cyber. This did not endear Cyber Promotions to the average Internet user. As a result, a concerted public relations attack was made on their provider (I believe it was Sprint) to try and get their net connection disconnected. And this was finally successful, but Sanford had a trick up his sleeve: AGIS. So AGIS connected CyberPromo, and they have been connected ever since. This has provoked numerous attacks, hate mail and crack attempts against the AGIS systems, some of which were successful. Since AGIS never really had the good people needed to stem these attacks, this was a real problem for them. However, so far AGIS has stuck to its guns, and the $ 6,500 a month per T1 revenues that CyberPromo is paying them. Hmm, it only takes $13k a month (two T1s worth) to bribe a backbone provider? Odd. AGIS has helped found a junk email "global remove" registry. However, so far it's been rather laughable. Just for fun, Justin Newton of Priori created an account on one of his machines and subscribed it to the "remove" list. Sure enough, he's been getting spam at the rate of 4-5 a day, thus showing that there may be another use for the "remove" list. When AGIS tried starting a morotorium on junk email while they were formulating their anti-spam plans, Cyber Promotions broke the morotorium by finding another connection, multi-homing and using the second connection so the spam could keep flowing. This upset even AGIS, and they cut Cyberpromo off for a brief time. But within 24-odd hours, he was back up, running and as unrepentant as ever. Service quality, as far as I can tell, has continued to deteriorate. I cannot recommend AGIS at this point. Before I wrote that summary, I had some earlier material Karl Denninger, who gave the venture much of its early reputation, recently sent a message to the Inet-Access mailing list giving the reasons he left Net-99. Apparently they told him he had a stake in the company based on a handshake deal, but then when Net-99 was sold, he was neither consulted nor paid. In addition, he believed that they were taking advantage of the equipment he had effectively donated to the effort; they promised him a Cisco 7000, but never delivered. As a result, he became disillusioned with the company and quietly severed all connections with it. An anonymous Net-99 customer had this to say about the merger: (begin quote) Some anonymous comments on AGIS/Net99, as I have been a customer since August. The transition has gone smoothly. AGIS is moving Net99's infrastructure from MFS Telecom over to WillTel. The moving of the circuits and routes has gone well. There have been some bumps in the billing department. Most of the errors have been in the customers' favor, so no complaints here. I have heard noise from former Net99 employees, not sure if it is griping or grounded in fact. I have heard some noise from AGIS about how Net99's infrastructure was cheap and poorly designed, etc. Sounds like someone upset somebody in the buyout. All in all, their performance hasn't changed much since the buyout. They've raised their prices, so I'm glad I got in when I did. (end quote) Still, Net-99 customers seem to be staying with the company, although some have griped about service problems. We'll wait and see what happens here; I'd like to hear more comments from Net-99/Aegis users for the current status of the venture. Your anomynity will be protected if you so request. Since I wrote that, people have started complaining vociferously about Net99 service quality, in particularly with regard to their reverse DNS. The reverse DNS problem has gotten so serious that an informal mailing list has been created for Net99 customers to discuss their complaints. For information on Net-99, check out http://www.net99.net/ . For AGIS, look at http://www.agis.com/ . +.# Where could I get a list of national and local providers? An excellent starting point is the DLIST, "an online list of Internet service providers who offer dedicated line connections." To find out how to receive an updated version of this list, send mail to dlist@ora.com. (From the book 'Connecting to the Internet: An O'Reilly Buyer's Guide', by Susan Estrada, which includes a printed copy of the DLIST in an appendix). Here is the ``Yahoo'' reference, which is apparently more up to date: http://www.yahoo.com/Business/Corporations/Internet_Access_Providers/ Note that this all has to be on one line. To scope out your competition before taking the plunge, an excellent resource is The List of providers at http://www.thelist.com/ . For quite a while, this service even provided ratings for the services from customers. Unfortunately, that part of the service had to be discontinued due to abuse. A great pity since it was an extremely useful service. +.# What about a SLIP connection? For the most part, a SLIP connection is not considered sufficient for a serious provider. However, it may be the best way to start if you are unsure of demand for your service or want to test-market your ideas. It certainly lets you hook up for a minimal amount of money, assuming you can find a resellable connection. If you get SLIP, try to get CSLIP (compressed SLIP), and make sure you know what baud rate the line is fixed at. I got a 28.8k SLIP with the baud rate fixed at 28.8, and the result was that I could not get a newsfeed consisting of alt.* plus rec.* without falling behind. Be sure you get as good a SLIP connecion as possible. + Should I join an Internet Trade Association? +.# What's all this about the ISP/C? The Internet Service Provider's consortium is a trade organization supporting the ISP industry. For amounts ranging from $ 100 to $ 1,900 a year, you get a large mix of services, ranging from lobbying in Washington to joint purchase agreements with a number of vendors. "The ISP industry needs a voice in Washington," said Deborah Howard, President of the Consortium. "We are constantly being attacked by entrenched interests ranging from telephone companies to people in the government itself. Join the ISP/C and gain a staunch ally in the challenges that face our industry over the decades to come." For additional information, visit http://www.ispc.org , or the European Mirror site at http://www.euro.ispc.org . The mailing list is isp-list@ispc.org . +.# What is the Commercial Internet Exchange (CIX)? This section of the FAQ is an intriguing history lesson on how a painfully divided organization attempted to hold up the Internet, and how it was discovered to be an Emperor without clothes. If you're not curious to hear a little history, you can probably skip this section entirely. If you've heard about the CIX and aren't sure what it is or what it can do, read on. It's January of 1995, and the CIX has mainly slipped out of the news. In autumn of 1994, it seemed that you couldn't read your mail without another mention of controversial CIX moves. Now, however, things seem to have quieted down a great deal. Here's how things looked, circa November-December 1994: The Commercial Internet Exchange (CIX) is many things to many people. To some, it is the heroic institution separating small ISPs from the oblivion of disconnectedness. To others, its $ 7,500 a year annual fee is restraint of trade, monopolization or worse. The CIX offers to do three basic things for you: (1) Any CIX member must agree to pass through the packets of any other CIX member, without fee. So if you want to talk to a site in England, and you're going through five or ten other sites on the way, they cannot charge you for this transport service, assuming all the sites are CIX members. (Karl Denninger tells me in response to this document that they are only obligated to connect through the CIX router and nowhere else, although you are presumably guaranteed at least your access to the CIX router). (2) The CIX operates a router, based in San Jose, CA. As part of your membership, you have to be connected to that router either directly or through a chain of CIX members. If you cannot connect to a specific site through normal means, you can use the CIX router as a last-resort option to get where you need to go. A direct connection to the router costs $ 5,000 on top of the membership charge. (3) Unspecified lobbying and public relations efforts. Observers on the inet-access mailing list state that these efforts are negligible. However, those people are also anti-CIX for other reasons, so you may wish to take their opinions with a grain of salt. The CIX has not issued any official commentary that I know of on the specifics of these efforts. There has been an enormous amount of talk - and we're talking about literally megabytes of stuff - about whether the CIX as it stands is a Good or Bad thing. Non-CIX members and some CIX members not on the board have made the following points: (1) The basic principles of the CIX are worth saving. Even the most virulent hater of the organization has stated that the principles of settlement-free peering are the glue that holds the Internet together - and those principles should be supported by a trade organization people join. Gordon Cook's remarks are typical: "CIX may well go away, but if it does Sprint and MCI in a year will have life and death power over the small fry ... etc ... ie what if they forbade resale?" Gordon is no friend of the CIX in its present state; he has made many scathing commentaries on the present situation. (2) The CIX is imposing route filtering, as of 15 November 1994. Previously, if you were a non-CIX member and were connected through a CIX site, you could still use the CIX router in the same way CIX members could. Under route filtering, the CIX router will become for the exclusive use of CIX members. Non-CIX members consider the $ 7,500 annual fee to be excessive; depending on the size of their operation, it could nearly double their operating costs. In September 1994, the membership had an annual meeting, at which they voted to NOT impose this route filtering. The CIX Board of Directors said that they would have to impose filtering in any case, for legal reasons. Some members were relying on the CIX connection to hook themselves to all providers, whether members or no, so they actively did NOT want the filtering to occur. At the end of the meeting, people like Karl Denninger were confident that filtering would not be imposed. However, on November 1, the CIX announced that, for legal reasons, the filtering would be imposed on 15 November. At that point, Karl Denninger's provider and Net-99, a joint venture of Karl Denninger and Joseph Stroup, made a very public resignation from the CIX. See below for some additional details). On 15 November, supposedly D-Day for filtering, someone on the Com-Priv mailing list noted that nothing had happened, and that even CIX routes were still available to all. Bob Collet admitted that the filtering was being implemented only gradually. Some people have said that filtering is impossible, given the specifics of route handling in the type of router being used by the CIX. (3) The CIX has a router, and some people are a long, long way from it. Wouldn't it be better to have lower membership rates than a router, especially since the CIX can be thought of in some sense as competing with its own members? (In fact, one of the CIX members - PSI - operates the CIX router under contract). All this sound and fury can be quite entertaining, unless your business depends on it. However, you may be able to relax: only 38 of 32,000 routes are unique to the CIX router. What this means is that if you're not a CIX member, route filtering will do very little for you; you lose contact with just 38 sites. If, however, you are a CIX member relying on the CIX router to connect you to the world, you may find yourself cut off from numerous non-CIX sites. As a result, most of the people connected through the CIX are apparently scrambling for alternative connections even as we speak. What seems to have happened is that connectivity providers such as Sprint and Net-99 are in practice taking care of routing for their customers. As a result, the CIX router has become very nearly worthless, especially since it's been heavily overloaded. Bob Collet's latest statement is that "a phased deactivation of the router" will occur, probably to conclude in late 1995. At that time, the CIX will then become a (cheaper) trade association which will continue working for the settlement free system as described at the beginning of this section. Other CIX Board members, however, are hurling insults at all who would dare to ask them questions. The following comments have been made by people on the Com-Priv mailing list on this subject: * If you're connected by a provider that has access to MAE-EAST, a major interconnect point, you'll be able to connect to people hooked in to all major providers, since they're all there. This, for example, should cover both Sprint and Net-99 customers well. * However, you will NOT be able to connect to people who are hooked in solely to the CIX router, unless you become a CIX member. Fortunately, very few sites appear to be in this category, and most of them are scrambling for alternative arrangements even as we speak. * Providers don't advertise all their interconnected routes to MAE-EAST; they only advertise those routes which help provide connectivity to their customers. For example, if PSI has a direct link to Hong Kong, and PSI is connected at MAE-EAST, you'll be able to hook up to all PSI customers, but not necessarily Hong Kong. However - just to confuse things - if the Hong Kong Supernet, a specific site, was connected through PSI, you would of course be able to hook up to it. As a result of this CIX filtering, Karl Denninger's MCS-NET and Net-99, his joint venture with Joseph Stroup, have resigned from the CIX effective immediately. Their complaint is that the obvious wishes of the membership (as voted on in the earlier meeting) were ignored. Bob Collet has asked them to suspend their resignations temporarily, while he attempts to put together a solution to please all parties. As a result, they have not yet put their resignations in writing, but their very deep displeasure with the CIX and its representatives could hardly be more clear. The CIX response has been that the filtering is a legal necessity, according to advice of counsel. Others say that the law could be read either way, and that the primary goal of the CIX is to expand connectivity; filtering is unquestionably not going to advance this goal. It's worth repeating the requirements for filtering, to give us a better understanding of what's going on. Filtering would affect you if: - The site that feeds you is not a CIX member - You are not a CIX member -AND- you sell SLIP/PPP access, or any other form of IP connectivity (56k lines, T1, etc). Otherwise, you can relax and ignore this issue completely. So if you sell shell or BBS accounts, and the site you're connected to is a CIX member, then you have full CIX connectivity and can cheerfully ignore this issue. Otherwise, the bottom line seems to be that you could ignore it anyway. Stay tuned, though: this optimistic vision is probably good for the next 30 days and no longer. Bob Collet (rcollet@sprint.net), a spokesperson for the CIX Board, was kind enough to review the above. As his conclusion to a short list of suggestions, he wrote: "Suggest toning down the personal opinion flavor of the document." His belief is that my anti-CIX prejudices are showing rather blatantly, and that an impartial document would be more neutral in tone. I have invited Mr Collet or any representative he wishes to choose to contribute a statement of CIX policy, which I will insert here verbatim if and when it arrives. Bob Washburn was Executive Director of the CIX until a month or two ago. Mr Washburn was taking care of most of the public relations for the organization, and his departure seems to have left the PR tasks up to members of the board. They are presently searching for a replacement, which may be one reason the flow of information out of the CIX is so sluggish. There are now approximately 100 CIX members. It is thus worth noting that this document consists of an analysis created by me using the best information I know - the various messages I read on the inet-access and com-priv mailing lists. While I hope it has been of value, clearly it cannot guarantee anyone's future policy, and does not constitute legal advice. I will definitely plead guilty to making an attempt to give this rather dry material a bit of entertainment value. +.# What is NET-99? (This section is largely obsolete, since Net-99 is effectively no more. See the previous section on national providers). According to Karl Denninger , one of the founding members: "Net99 is a commercial effort by Joe Stroup and myself to provide a resale-encouraged, peering-mandatory, backbone environment for the small and medium sized reseller. Net99 is not a cooperative, or a non-profit. Net99 is, however, an alternative project which should, if preliminary indications are good, provide a reasonable and affordable alternative for the reseller of Internet connectivity when squared off against the large companies now domainating this portion of the network infrastructure." Net-99 beat their November 1 deadline by a very considerable margin, having POPs up in mid-October. They presently offer service in the following cities: New York, Houston, Los Angeles, Chicago, San Jose, Boston, Philadelphia When asked for the number of sites presently on Net-99, Karl said, "Can't give you an accurate count; its growing incredibly rapidly!" Robert Gibson writes us as follows: "We have a connection via T1 to NET99, and it works very well. We also have other T1 connections into the Internet via Sprint, Navy. I have found the connectivity and support GREAT. ... NET99 is *great* in terms of service, and I would gladly pay a few $$$ for service, and enough $$$ to keep the network growing." Other Net-99 customers have been similarly effusive; I have yet to hear a single complaint, which is quite rare on the net. +.# What is Metering, and why are people so emotional about it? Metering is the idea that users and service providers should be charged depending on how much use they make of the service. At first blush, it seems only fair; if you're on the net for an hour a week, you should pay less than those who are on for five hours a day. For customers of metered services, however, it tends to be a very bad idea indeed. There are really two types of metering: Usage-based charging of users, and usage-based charges made by backbone providers. Very few people have much against the former, since free competition seems to be eliminating it (other than the toll-free number options of many vendors, which have to charge by the hour to pay the phone company's cut). The latter, however, is a different kettle of fish. Most of the time, when people talk about metering, they are talking about charges per packet for Internet use. As a provider, you would be charged for backbone use by your connection provider (such as SprintNet). The problem, of course, is that this means you'd have to pass those charges to your users in the form of hourly fees; this would effectively eliminate the flat-rate pricing model that's been so successful. What's the argument for flat-rate pricing? There are really three: People are much more eager to use a resource when it's not metered. Most people won't use a service for anything but the most vital needs when they hear the clock ticking in their heads. The net's ethic is founded on volenteer work of all kinds, like the production of FAQs such as this one, the moderation of newsgroups, and so on. This work would be prohibitively expensive with metered use, and the amount of information available would be much lower. Second, flat-rate is almost always cheaper for the user than the metered option. Consider two pricing models I've seen: $ 6/month plus $ 2/hour (UUNET) $ 17.50/month flat rate (Netcom) If you spend just 6 hours on the net each month, the flat rate becomes cheaper than the metered price. If you really enjoy the net and spent 3 hours a day on the system, your bill would be nearly $ 200 a month! Finally, measured accounts can be a real hassle due to the difficulty of administrating the timed accounts. This administration costs a remarkably high percentage of the gain in revenues obtained. The Internet has become the lively, fun and often bizarre resource it is today largely BECAUSE people didn't have to pay by the hour or by the packet. If the big communications near-monopolies ganged together and offered only hourly rates, the Internet would be changed beyond recognition. This is why people who are normally the most ardent free market advocates are pushing for some kind of regulation of backbone services. +.# What is a Firewall, and do I need one? Sort answer: If you are an Internet provider, you almost certainly don't need or want a firewall. Long answer: A firewall is a machine that separates your internal company network from the wild and wooly Internet. Ideally, it will let employees of your company do things such as send mail and browse the web without exposing your systems to the security risks that normally would come with such. Here's a simplified diagram: Internet | Router | Firewall | Your machines Note that all packets coming in from the Internet go through the firewall. The strictest form of firewall doesn't pass any packets between it and your machines at all; mail, for example, will run on the firewall machine and stay there until it is called for by one of your company machines. Packets thus cannot go directly between your company's machines (in the internal network) and the Internet. The problem with this, of course, is that your machines cannot perform useful work, either. You cannot surf the WWW or FTP files from your PC, since packets won't go through the firewall. A program called a Proxy Server can help with this. A proxy server takes HTTP requests from your internal network and sends them along to the outside world; then it takes the responses and returns them to your internal network. As a special bonus, this server can also cache (save) pages to the firewall system's local disk drive; then later requests can be fulfulled instantly, without the delay of full transfer via the Internet. (This is what the Prodigy and AOL browsers do). If you're an Internet provider, of course, your users will expect full and direct access to the Internet. Thus, you really cannot use a firewall machine for an Internet provider. For detailed information on firewalls and other security considerations, check out this URL: http://www.telstra.com.au/info/security.html + Personnel and Hiring +.# How do I find clueful employees? Good question. Sending employement offers to the ISP mailing list is thought of as being in poor taste. Asking your technically related friends and hanging out on the ISP mailing lists and absorbing the culture are probably the two most important things you need in your search. Maybe someone who's actually hired someone can help me with this section? +.# How do I keep them? Offer stock options, free lunches and plenty of free soft drinks. Those perks are amazingly popular, especially if you can't offer much in the way of pay. My advice is to always be nice to your employees, and they'll love you back. Even Generation X can be co-opted; you just have to make the effort. Most companies, of course, don't, so if you can, you have a built-in advantage that's quite difficult to overcome. +.# How many technical support people do I need? Technical support is the bane of the industry, since your people are encountering angry and frustrated customers. Your employees won't like doing it. According to Avi Freedman of Net Access, "I have seen 1 person + 1 person/1000 users + 1 person/100-200 users signed up/month, depending on how clueful the base is (advertising vs. referral) and on how good your setup software is." This is a generic estimate that comes from his own personal experience and that of other ISPs; it's not solely a Net Access answer. The alt.sysadmin.recovery newsgroup contains messages that will give you some insights into the rather painful issue of technical support. + Internet Software tips, tricks and answers After looking at all these incredibly high costs associated with getting on the Internet, it is somehow a relief to venture into the world of software, where virtually everything you would need to use is available at no charge other than a few hours of pain setting it all up. There is a massive amount of information on the net on Internet software and how to set it up. The O'Reilly zoo, a collection of books for Unix system adminstrators with pictures of animals on the cover, is strongly recommended by most people who've read them. I have a pretty healthy zoo, and fortunately the cover critters haven't chewed through my cables just yet. (I keep a close watch on them, though). Because much of this stuff is well known to anyone dropping by the bookstore or reading their manuals, I'm only going to mention some tips I've gotten through (sometimes hard) personal experience. +.# The Permissions headache Perhaps the hardest thing to get right when setting up your provider software is Unix permissions. The permissions system is an indispensable way to straighten out who can do what on your system, but it becomes absurdly tough to manage certain programs. I managed to lose three weeks worth of news while vainly attempting to straighten out some particulary ferocious problems of this type, although this was mainly because I was unlucky enough to have left town before checking that one simple change I'd made would hold up. Don't repeat this mistake - always check permissions whenever you make any change, however minor they may seem at the time. Kevin Kadow reminds us that you should use tight permissions for security. One provider who offered him a position had been running for the last year with their /etc/motd set to world write access, so any stray user who figured it out could change it to whatever he wanted. "The basic rules for secure Unix permission (as I was quoted in PC Week on) are to provide the minimum permissions necessary. Never give world write access to a file or directory when group write access (or better, user-write) would be sufficient. Never make a file set-group-id (or horrors, setuid) unless it's totally unavoidable. +.# Which news software should I run? This is an interesting question, since most of the books and other documentation on news software are way out of date - particulary the O'Reilly/Todino book that is so often described as the definitive guide. At the time that work was written, C-News was the classic, recommended release of news software. Because of that book's recommendation, I spent more time fighting with the incredibly slow C-News unbatching software than I care to remember. It would virtually freeze up my system whenever news was being unloaded! As soon as I switched to a SLIP, I switched to InterNetNews (INN). What a difference! My system does not slow down while INN is receiving news. At all. If you need to accept full newsfeeds, I highly recommend INN. For other options, see below. I ran my INN on a Linux system with 20MB RAM. (I now have a 32MB system). Many Sun users think a reasonable minimum for running a full newsfeed is around 32-48MB. INN is definitely a memory eating machine. I now use my provider's news server, so I'm somewhat distanced from news issues nowadays. Unfortunately, the hardware requirements to run news have increased dramatically since I first wrote this document, because news volume has gone from enormous to absurd. For this reason, I strongly recommend that the new ISP forget about running news at all, and get it from other people's servers. See below for additional information on this option. If you still need to run news, I've been hearing good reports about a program called DNEWS, a commercial substitute for INN that costs $ 495. Tim Perry writes: "It is worth the price. My news headaches went from daily to almost nil. I have yet have anyone say they couldn't read news. I have three full feeds coming in, one going out and before with INN running a smaller load I had a load avg of 3.5 normal run and expire run up to 5. With dnews, I get an average run of 2.0 and an expire at about 3.0. Also, the expire time went from 5 hours to about 30-45 minutes. Pretty good I say." DNEWS works only with NNTP-based readers, which means that if you have old-style software that tries to read directly off news spool directories, it won't work for you. This would really apply only if you sold shell accounts and didn't update your news reading software; all reasonably modern software, whether under Unix or Windows, will work. DNEWS is available for VMS, Linux, Solaris [SunOS 5.x] and (choke) Windows NT. You can find out more about it at http://world.std.com/~netwin/ . DNEWS has two interesting advantages over INN: (1) It can do a "sucking" feed, which is more efficient and uses less disk space, and (2) it stores news in clumps of 100 messages, instead of using one file per message as INN does. This saves i-nodes and eliminates a lot of the disk-grinding horror produced by INN and other conventional news systems. DNEWS' authors appear to be emphasizing the Windows NT platform, which makes me feel intense nausea. See the above discussion of hardware platforms for why. Nonetheless, the fact that someone's actually put some thought into designing a news server is pleasing. There's been a lot of talk on various newsgroups about doing a more sophisticated news server that eliminates the problems with the one file per message system; the authors of DNEWS are the first people who have actually done it. Good for them for doing that; bad for them for supporting junky Microsoft products. If you must use an Intel platform, use their Linux version instead. :-) +.# Let's back up for a minute. What IS news, anyway? News is many things to many people. To start with, it's not really news at all. The closest analogy is a gigantic distributed public message system with its own customs, folkways and lore. Generally, the more intelligent Internet users read and write news, leaving Internet Relay Chat (IRC) to the dull (see below). News is available on over 13,500 different topics, which represent just about everything there is in our universe - and even in many universes yet to be discovered. Just to complicate things and make them interesting to the unwary, News is not Internet News; it's USENET News. What this means is that more primitive systems with only UUCP (Unix to Unix Copy) access can participate in news. Other Internet services, with the notable exception of electronic mail, require a direct Internet connection. The main problems with news relate to its sheer bulk. Every day, between 40-450MB of stuff flows into your system (depending, of course, on how large a newsfeed you get). So that your disks don't fill up constantly, the news software "expires" it automatically; that means that old messages are automatically erased to make room for the new. On my 486/66 running Linux, the load average zooms up from .63 to about 3.78; the disk starts whirling like a banshee, and programs that normally load instantly are suddenly faced with catatonia. As a result, you'd better do your expiration when nobody else is using the system - like around 3am. Unfortunately, on some systems it can take take up to 12 hours for expire processes to run, meaning that there is no time your server is not being eaten up by this process. Usually this means you should get a dedicated news machine, quick. When I asked Craig Warner of CERAM, the Sun clone vendor, whether news would perform well on the shiny new Sun clone I'm contemplating, instead of my 486/66, he said: "The design of USENET news is a big problem on all systems. In general, the performance of things like expiration is determined by the access time of the disk drives. This is where having large drives significantly DECREASES performance. "To minimize the impact, news should ideally be on a separate machine, and the news tree spread out over multiple physical disks. if you do keep it on the same machine as your interactive accounts, the news disk(s) should be isolated from the operating system and the user directories. "The ideal solution for a news tree is to keep it on a solid-state disk - but this is way too expensive for most sites." I'd say a close second would be to run your INN on a separate machine, used for nothing else. If you are using certain versions of Unix, including Linux, you may have a major problem with news: Your system may run out of i-nodes long before actual disk space is exhausted. SunOS solves this problem by normally allocating enough i-nodes for a disk consisting 100% of 2k files. Other operating systems, including Linux, get more indicated disk space out of a given drive by allocating an i-node for every 4k of space. This is fine - better, even - for every application but news. You should do a 'man mkfs' on your system to find out exactly what you need to do to change the block size for your news disk before formatting it. (If you're not familiar with the concept of i-nodes, note that there has to be one for each file on your disk). Andrew Hadenfeldt added some additional facts. - It's worth checking out your system's default block size. On many systems, such as IBM's AIX, this size is 4k. A 2k block size will result in much less wasted disk space on your news partitions. - Daily cleanup processes, such as expire, need spool directory space, too. Be sure to leave at least one day of spool free as workspace for new incoming articles and old expiring articles. - If you plan on feeding other sites, be sure to allocate space for this as well, especially if they're using low-speed connections and likely to run a backlog. +.# Where does News come from? A USENET newsfeed normally comes free of charge from the people who set up your Internet connection. If you cannot get it from them for some reason, your fellow ISPs can provide the service for fees ranging from $ 50 - 75/month. A full newsfeed currently runs around 1GB per day, depending on how good your feeding site is. News has become much less reliable of late, due to the massive and ever-increasing volumes. Because of this, the best news sites take three or four different news feeds to ensure the best possibly distribution. One particularly interesting option is pagesat [ http://www.pagesat.net ]. They use a satellite system to deliver you a full newsfeed. This is great because it doesn't use too much of your bandwidth. Most users think service is poor, but at the same time appreciate its ability to supply a newsfeed that simply could not be managed with their bandwith alone. Service quality can totally collapse, by no fault of PageSat itself, when the atmospheric conditions are unfavourable - i.e. during heavy rains or other adverse weather conditions. As a result, good quality backup service is still required for news even if you decide to obtain PageSat service. Unfortunately, as of December 4, 1996, PageSat appears to be out of business. Their CEO, Norman Gillespie, claims that he is putting something together to continue the service; however, various claims of flakiness and poor service make me wonder very seriously about the quality of his offerings. Another service called Planet Connect [ http://www.planetc.com ] has been around for quite a while to offer Fido news and USENET. Users have commented on the high level of attention they give Fido and the relatively low quality of their USENET service. Perhaps with PageSat's demise, they will scent opportunity and improve things. Or perhaps not. If anyone does try Planet Connect, I'd like to hear about it. +.# Which mail transport agent should I run? At this point, probably qmail. Qmail is an up and coming program with relatively simple configuration, reliable operation and high-quality support. We're using it here on Freelink, and I'm extremely happy with its performance so far. Visit http://www.qmail.org for more information. Information about spam, spam prevention, etc, is in the later section on dealing with users. Despite its fearsome reputation, sendmail still has many advocates. For one thing, a critical mass of people use it, so there are a lot of recipes and workarounds for you to play with. Despite what you may have heard, if you arm yourself with The Bat, aka the book "Sendmail" by Bryan Costales, Eric Allman and Neil Rickert, it's not too hard to modify an existing sendmail configuration file for your site. The tutorial walks you through all the various commands and the steps you have to take to customize it as needed. There's surprisingly little you really need to do or understand to set up a basic file. Eric S Raymond was kind enough to give me a sizable amount of information about sendmail and its chief rival, smail. Sendmail presently wins over its rival by allowing multiple transactions (that is, posts of mail) per SMTP connect; smail is not so generous. This means massively poorer efficiency for smail and thus his present use of sendmail. Eric says: "This makes them significantly less efficient than sendmail for sites with high mail volume. When this changes, CCIL [Eric's system] will move to Smail 3.0 and drop sendmail *instantly*." As a BSD user, Eric has this to say about the task of sendmail installation: "The stock sendmail V8 supplied with BSD/386 1.1 and up works pretty well. The V8 developers did a fair job of hiding the mind-wrenching ugliness of sendmail config files behind a layer of more civilized m4 macros. Note: be sure you install the CERT security patches for V8, or that your vendor has done so." Tony Sanders adds: "You should mention that people want to use sendmail V8.6.9 (or better) and that it's available from ftp.cs.berkekey.edu:ucb/sendmail." This includes the CERT security patches, so you don't need to worry about them. As a Linux user, here's my experience: I used sendmail IDA, which is available (from sunsite.unc.edu) pre-configured for Linux users. This, naturally, makes IDA the logical release for that system. I attempted to compile Berkley Sendmail V8 on my Linux box, but found it impossible without the installation of massive amounts of additional software. As a result, I decided to give up on V8 and use IDA. I found The Bat to be a faithful friend once I got to know it. After reading appropriate parts of The Bat, I was even able to write my own "Mail Delivery Agent" to convert standard network mail into BBS format mail (stored on my system as consecutive text files, just like news). It was easy and even fun; my opinion of Sendmail improved dramatically when I read the parts of The Bat that effectively took me by the hand and told me how that could work. Both Eric and David W Rankin Jr have about the same opinion of smail. They tell me that it's relatively easy to configure, particulary for mailing lists. Unfortunately, it's hampered, as said earlier, by mediocre SMTP support. David Rankin also notes that it's not well suited to complex hub/spoke routing schemes. As a result, you might well wind up getting smail while it's suitable for your small site, and then having to configure sendmail or another more powerful program later. David W Rankin Jr also puts in his recommendation of MMDF, an older mail transport agent which is now not too well known. "In case you're not familiar with it, MMDF is a mail transport system (like sendmail or smail) designed for higher traffic sites that do multiple protocols or lots of routing. MMDF is older than sendmail, generally more secure than sendmail (it uses several programs, isolating the root parts. Of course MMDF has been pounded less than sendmail), and a LOT easier to configure (IMHO)." His present site uses MMDF, as will any sites he administrates. However, he has not yet configured MMDF; he has only seen it successfully in action. He says, "take all I say about it with a shaker of salt :)". MMDF can be obtained via FTP from a.cs.okstate.edu. Finally, there are now commercial mail programs, such as post.office and Microsoft Mail for those unfortunates among you who run NT. post.office is praised by many for its ease of configuration, but roundly cursed for poor support - the post.office user is still pretty helpless against the spam onslaught. post.office is also mind-wrenchingly expensive for large sites. +.# How do you set up a "Phantom Domain"? Many users want a package of benefits from you that will include an indentification of their site as theirname.com. This makes it look like they own a site on the Internet, which is a sign of some prestige in the net.world. To make mail to user@theirname.com work, you can hack Sendmail as described below. My thanks to Christopher X. Candreva for passing on this information. Note that this is quite a bit easier in qmail. (Beginning of quoted message) From: rsanders@interbev.mindspring.com Date: Wed, 15 Feb 1995 18:58:09 -0500 Subject: phantom domains/fake mail ids I don't recommend that people just configure their machines to accept mail for foreign domains as if it were local; it's inevitable that your customers want info@whatever.com and sales@, etc. Furthermore, I don't modify my sendmail.cf anymore; I use sendmail V8's ability to use an external DB file to do lookups. I only had to write four rules to do that. ----------------------------------------------------- S98 R$+ < @ $+ . > $: $1 < @ $2 > . R$+ < @ $+ > $* $: $(maildomains $1@$2 $: $1 < @ $2 > $3 $) R$+ < @ $+ > $* $: $(maildomains $2 $: $1 < @ $2 > $3 $) R$+ < @ $+ > . $: $1 < @ $2 . > ----------------------------------------------------- With a DB file that holds these associations: ----------------------------------------------------- jdoe@someplace.com jdoe janedoe@someplace.com jane someplace.com spowner ----------------------------------------------------- I can have mail to jdoe@someplace.com go to the jdoe mailbox, janedoe@someplace.com go to jane, and all other mail be delivered to the "spowner" mailbox. It's not a bad setup, although I still want to tweak it to short-circuit that DB lookup altogether if the destination domain is not present in a certain class. This sort of thing is laughably easy if you have the O'Reilly sendmail book. I recommend it. -- Robert (end of quoted message) +.# What about Domain Name Service (DNS) This is one of the most frankly confusing aspects of setting up a provider. I've been able to dodge the issue by letting my SLIP provider handle it. Dave Van Allen quite correctly chides me for the omission: "The other area that seems to be missing is a mention of the importance of Domain Name Service. DNS can be a major performance player in your server calculations, and to be in-check with the NIC, a domain is required to have two name servers running on its behalf. One of the most complicated operational areas a new ISP will face is configuring a DNS. (The other will be the dreaded sendmail.cf). "The Cricket Lu book [O'Reilly & Associates] on DNS is very good and a must-read for anyone venturing into this territory." I believe your Internet wholesale provider normally supplies the two DNS servers required by the NIC. However, I believe that just goes to the top of your domain; if you have multiple machines on a network (as almost all providers will), you will need to run your own name servers for them. Kevin Kadow quite rightly corrects this section of the FAQ: It's not that hard to set up DNS once you get the rather confusing and convoluted syntax right. "Even if you are not the DNS for any domains, you'll want to have a local caching DNS for performance reasons. All UNIX operating systems generally include the required 'named' software to perform this function." I was able to set up DNS on my Sun SPARC system in less than a day, using the instructions in the O'Reilly book 'DNS and Bind'. Get the book and you shouldn't have much trouble. Kevin's company offers DNS service and mail forwarding for $ 49 a year. If you need a DNS server for your domain name applications, and you're still searching for a wholesale provider, this is an indispensible service. You can contact him at kkadow@msg.net. +.# How do I set up a Gopher server? The Gopher server is proprietary property of the University of Minnesota and is available to the commercial world only under rather stiff terms. "Commercial use" requires a $ 500/year payment, regardless of the size of your site. If you use it in a direct order taking application, they want a percentage of profits. Gnu is apparently coming out with a free gopher server, available from: ftp://ftp.acns.nwu.edu/pub/gn/gn-1.0.tar.Z The newsgroup comp.infosystems.gopher covers gopher more fully. For the most part, World Wide Web (WWW) seems to be replacing Gopher as the information distribution tool of choice. Gopher clients seem especially vunerable to running away when the user hangs up on them, using incredible gobs of CPU time. This is also true of Internet Relay Chat (IRC); see the IRC section. I have included my perl script for handling this problem in the technical problems section. Gopher does have one significant advantage over WWW, however: its more formally organized nature makes it easier for new users to understand. +.# Secure Servers and SSL We all know the Internet is filled with dragons. Evil dragons, with but a single eye, spying on your network, searching for your passwords and credit card numbers so they can find out your most intimate secrets. Or, better yet, charge you for their upcoming trip to the Caymans. The fear of these dragons has been played up by the media, who had a massive stake in making people frightened, and Netscape, which wanted to sell its new secure server software. And so we all fear the dragons, and we look to Netscape and friends for solace. So you buy your Netscape server - outrageously priced at $ 5,000 a pop until recently; now a more reasonable $ 295 to $ 1,000 - and install it on your network. And then you learn you need something else: A certificate that identifies yourself to your customers. Basically, the idea is that, since Netscape's reputation is on the line whenever the solid key of a "Secure Server" appears in your browser, they want a trusted company to verify that you, the owner of the server, have the right to collect money on behalf of your company. Because of this, they have appointed special companies called Certificate Authorities (CAs) to supply a secure certificate that verifies that you, sa the server owner, actually have a right to collect money. Note that this is required whether you're planning to collect money or just make your documents more private. Unfortunately, these certificates are quite expensive. The leader in this field - and the only company that Netscape 1.x will accept and that Netscape 2.x will accept by default - is Verisign [ http://www.verisign.com ]. This costs $ 295 per year for the first certificate, plus $ 95 per year for each additional certificate. You need a different certificate for each web server you run, and for each secure web server program you use. So if you have a secure server at http://www.freelink.net/ and http://www.flashemail.com/ , as I will, you need to have two certificates. And if you change from Netscape to Apache web servers, you need to get a new certificate. Frankly, it seems like a big ripoff to me. But if you don't want pre-3.0 versions of Netscape not to scream at you, you have to get one from Verisign. A new company called Thwaite also sells certificates, at a flat $ 100 (soon to be raised to $ 125) each. However, they will only work on Netscape 3.0 and later or MSIE 3.0 and later. As a result, I find it hard to recommend them; you're shutting out a healthy percentage of your market. Note that the certificates included with new copies of Apache are Thwaite. If you use a certificate unrecognized by any Netscape version, it will give a warning and then continue into your site. If you use a certificate unrecognized by MSIE, it will simply refuse to proceed. (This is due to the integration of the certificate system with their ActiveX control technology, whose "security" system relies strictly on certificates). So unfortunately, if you need to accept credit cards on the web, you probably want to stick with a Verisign certificate. +.# What about running a World Wide Web Server? The World Wide Web is perhaps the most important of the Internet services, because it attracts people like bees to honey. Even Forbes magazine, in their January 30, 1995 article debunking Internet marketing myths, couldn't resist putting some of the more glitzy and creative WWW efforts in their article. (Oddly enough, they highlighted some of the worst ones, because their bandwidth-killing images looked truly splendid in print.) What this means to you is that there's a potential bonzanza available in Internet marketing. See the section on Internet Marketing, below, for more information. I have set up a WWW server, from which you can get the latest version of this FAQ, as well as other bits of useful related information. Check out http://www.amazing.com/ . Of all the network utilities I've used, the NCSA WWW server has been by far the easiest program to install. I downloaded it, compiled it, and it worked. Simple. The newer Apache server was very similar in ease of configuration, and has been rock-solid reliable on both Sun and SGI platforms. I have also tried the Netscape FastTrack and Commerce servers. They are a snap to set up even if you're a GUI lover, but seem to burden a machine quite a bit more than Apache does. Learning HTML isn't too hard, at least for the simple stuff. It's not even that hard to write utilities that convert text to HTML, complete with automated sectioning and tables of contents; that's how the HTML version of this FAQ was prepared. HTML will operate pretty well, even on a 28.8 SLIP connection, if you limit the size and number of images you put on your pages. Remember, a picture may be worth a thousand words, but it's also bigger than a thousand words. Quite a bit bigger! There was originally a great internet marketing list at popco.com, but unfortunately the moderator gave it up. Its function is now split between several lists: webdesign@list.webmonster.net covers the complexities of web design, with a large number of really hard-core design types on the list. It's one of my favourites. Visit http://www.webmonster.net/lists/ to subscribe. websociology@list.webmonster.net discusses ... well, just about anything. It's a lot of fun - if you're interested in social issues as they apply to the web, you should definitely subscribe. Again, visit http://www.webmonster.net/lists/ for details. market-l@telmar.com is a wide-ranging and chatty discussion of marketing. Intriguing. Depending on the nature of your users and what you allow them to offer in their personal WWW pages, your server may be subject to some truly awesome loads. Be sure to take this into account when determining fees for these services. A reasonable compromise might be to allow people to put their data on the Web at no charge as long as they used no pictures, and charge for any pictures presented in some way. The newsgroup comp.infosystems.www covers the World Wide Web. Some BBSs and small Internet providers are allowing people to create single-page Web pages dirt cheap. When selling your services, you might want to emphasize that a single-page web site is no fun at all. For more complete information on the World Wide Web, Tony Sanders suggests the following URL: http://www.bsdi.com/server/doc/web-info.html Kevin Kadow has a tutorial on designing web sites at http://www.msg.net/tutorial/ . You can find out more about the NCSA server at http://hoohoo.ncsa.uiuc.edu . In addition, information on Netscape and their extensions to HTML is available at http://home.netscape.com/ . +.# How do I choose a WWW server? There are four basic WWW servers available for a Unix system: (1) NCSA HTTPD. This is the original server, now greatly improved with multiple forking for fast connections. In testing, it was actually slightly faster than the Netscape servers. The multi-forking feature apparently has trouble working on Linux systems; it will work on BSD systems just fine, and is a major speed improvement over older servers. (2) Netscape Servers. The FastTrack server is meant for relatively low-volume serving; the Enterprise Server is their higher volume software. I'm not sure if there really is much difference between the two other than price tag. One of the few advantages of Microsoft entering the Internet market is the reduced price of secure servers; the Netscape Commerce Server was at one time a $ 5,000 product! Netscape servers are all fast and well designed, although not quite as fast as the current Apache server. (3) CERN HTTPD. CERN is mainly of interest to you if you're using a firewall. It can run on your firewall machine and provide access to your internal users by passing on HTTP packets through to your system. (4) Apache. This is a heavily patched version of the NCSA HTTPD, hence the name. Its main advantage is improved support for multi-homed hosts (see the next question). It should now outperform virtually all web servers. +.# How do I set up http://www.massive-company.com/ ? This is a question that comes up more and more nowadays: How do you set up a 'multi-homed' server, so you can support more than one domain on a single computer? That is, how can you make http://yourisp.com/ and http://massiveco.com/ be the same server? The short answer is that you have to register two different domains with two different IP addresses. Then, you set up your basic network support software to support more than one domain, and you set up your HTTPD daemon to recognize the multiple addresses, and point them at different data locations. The support for one domain is very easy under SGI Irix, the operating system I currently use. Just type ifconfig -alias www.massiveco.com 207.151.18.30 substituting massiveco's IP for 207.151.18.30, of course. Most modern operating systems have similar options for ifconfig. For SunOS 4.x, you'll need to search the web for the VIF (virtual interface) module. Some people will try to suggest software virtual domains; until people stop using Netscape 2.x or older, I don't recommend the practice. An excellent paper by Alan Barrett has been written on this subject; you can find it at http://www.isoc.org/HMP/PAPER/131/ . Many people find the new Apache HTTPD server to be the easiest one to set up for muli-homed applications; others note that it's currently slower than the newest NCSA. (This is because it's based on patches to the older NCSA server). The developers promise to speed it up soon. See the section above, "How do you set up a phantom domain?" for information on setting up mail for your new domain. +.# All About Web Browsers You will have to offer a SLIP/PPP account to enable the use of any graphical Web browser (other than SLIPKnot). See the section on SLIP/PPP accounts for full details. Mosaic was the first WWW browser to display both images and text simutaneously. It has now been largely replaced by Netscape (see below), which alleviates some of the speed issues that make use of Mosaic relatively tedious, and includes extensions dearly loved by many WWW page designers. There is a very real irony about Mosaic and Netscape: They attract users, because people want to see the snazzy graphics on the Web. But they also repel users, because many sites use large graphics and image maps that take forever to load. Because of this, about 40% of graphical browser users actually run with images off, according to someone's analysis of their log files. (Alas, I forgot who this someone was. :-( ). You might think of Mosaic and Netscape as tools ahead of their time, because the most popular uses have the hidden poison of slowness in them. For this reason, many people are looking forward to ISDN, which promises to provide the man on the street a 56k or higher connection. However, I doubt that this is a panacea; the links between machines are also heavily overburdened, and unless that's cured, ISDN won't speed things up much. Eric Raymond writes: Another problem with Mosaic is that it requires the Mosaic libraries to build. This means you either have to pick it up in binary form or pay about $140 for a Motif license. I couldn't get the binaries to work under Linux and I'm not going to pay $140 for something I'm just experimenting with, so I found an alternative; a package called Chimera, that works much like Mosaic but uses only the Athena widgets. You can find it on WWW's client-software page. Lynx, a VT-100 oriented WWW client, works very well if you don't absolutely need graphics. The graphics can be automatically downloaded to the client system if desired. It's worth noting that many Web users disable the graphics because they come up very slowly on a typical 14.4kbps SLIP connection. I have now used Lynx to download graphics I wanted to see, and it really does work quite well. Eric Raymond seconds this. He says "A lot of people poor-mouth Lynx, but it's pretty useful. At CCIL, all WWW access is through Lynx, and our users aren't complaining. That goes for the ones with Mosaic experience, too. What good is snazzy graphics if you have to wait a decade for a simple little logo to download?" A newer browser, called Netscape, solves many of Mosaic's problems and brings up a few new ones. Netscape's main advantage is that it loads all of the image and data associated with a URL concurrently. Because of this, you can read the text as it comes through and see parts of the pictures, even if not everything has yet come through. This is an incredible timesaver and makes Netscape the clear browser of choice from the consumer point of view. Netscape does have a few irritating problems, though. If you don't buy their multi-thousand dollar server system, it will give users a warning that they are not using the Netscape secure transmission scheme every time they try to transmit something on a form. This strikes me as a rather cruel form of extortion, even though the software is undeniably clever and well done. SlipKnot, a graphical browser running under Windows that's based on using Lynx from the Unix shell comand line, is an exceptionally clever solution that I think will increase in popularity. It's much easier to set up than genuine SLIP, and it's not much slower. +.# The Netscape Extensions and Designing HTML Netscape is now best known for its proprietary extensions to HTML. The most popular are centering, font sizes, backgrounds and tables. Unfortunately, the use of many of these features can create pages that are virtually unreadable on any other browser. Although it's true that Netscape now has an overwhelming market share, 25% of net users still use another browser - often Lynx, the non-graphical browser, and Mosaic, the original. Some people say that the Netscape extensions are so important that speaking to those 25% is practically like trying to speak to people with a different language. I must respectfully disagree with these people; it is possible to create a good looking page that works with all browsers, even though it will look best with Netscape. Because I have been working more on the web in recent months, I have come up with more up to date information on the web. Check out my Web FAQs: http://cgi.amazing.com/web-faq/ . Here are some comments on the Netscape extensions: > Image Maps Image maps are not a Netscape extension, but this seems like the best place to cover them. Most image maps are large images that take forever to load. ALWAYS make sure there is another way to get around your site, or many people will just turn back at the door. I would strongly recommend not using them at all; if your client cannot resist the pizazz, make sure they know that an alternate method of navigation is still needed. > Backgrounds and Colours Traditionally, the Web has used a solid, single-colour black or grey background. Naturally, people from the worlds of advertising and print publishing found this very drab and dull, and so Netscape created backgrounds to please them. In my explorations of the Web, I've found lots of people who use backgrounds on their sites, and they can have a pleasing appearance. At first. The problem is that most people don't seem to realize how much harder it is to read text superimposed on backgrounds. I used my IBM ThinkPad 750C computer with a NEC 5FGe monitor to view web pages; this monitor is far superior to the unit on most consumer PCs. Despite this, I had to strain my eyes in order to read the overwhelming majority of pages I sampled with backgrounds. If you must use backgrounds, please be sure to render the text in a contrasting colour. Netscape's colour setting features fit hand in glove with backgrounds. You should view your document on several different computers with different monitors before settling on a background and a colour combination. You'll probably want to use a nice big high resolution monitor for development, but make sure you keep an ancient 386 clone with a .39 dot pitch monitor around, and test your pages frequently with it. Remember, it's the only computer and monitor a lot of your customers will have. Never release a page with images until you have done this! The previous text was written in 1995 or thereabouts. Now, in late 1996, many sites have gone straight back to plain white backgrounds for readability. The use of anything but the most subtle background image has become reserved for either amateurish or arty sites. > Tables In the past, tables were probematical because they were only readable using the Netscape browser. At this point, virtually all users - including Lynxites - can read tables - and a modern, acceptable web page cannot be made without them. So use them as you wish. > Blink The Blink tag is largely discredited. For one thing, it looks unspeakably ugly on a Windows machine; the blinking is erratic. For another thing, blinking is just plain bad for your eyes. :-( > Center This is a basically harmless tag; if your browser doesn't support it, things will just all be left aligned. Not the end of the world for your readers. I recommend the use of this tag. There is a big flame war between those who want to use <p align = center> as versus <center>. The former is more standard; the latter is easier for humans to read. > Font sizes Again, I think of this as pretty much harmless; when it's not recognized, the document is rendered in a single font and size and is still readable. I have started to use Font Size tags on my own pages, mainly to create better-looking headings. If you do this, I recommend that you use this procedure:

This is a heading

This will show attractively in Netscape, but lesser browsers will still notice the tag and create a properly rendered heading. If you don't do this, you will find that your pages can be nearly unreadable on non-Netscape browsers. The main problem with Netscape extensions is that people who use other browsers - even earlier versions of Netscape! - may find it nearly impossible to read your pages. My best advice about the use of netscape extensions is to make your page as pretty as possible using Netscape, and then view it on other browsers and adjust it as appropriate. With a little extra work, you should be able to make your page look good no matter how people view it. HTML is now growing up with the new HTML 3.0 spec, which implements codes somewhat different from the Netscape extensions. Since the new versions of Netscape also support HTML 3.0, it is preferable to code your page using the new spec whenever possible; then it should work on all current browsers, not just Netscape. +.# What about Internet Relay Chat (IRC)? Internet Relay Chat is probably the most brain-dead use of the Internet short of downloading X-rated pictures from the binaries groups. Of all the things you will offer on your system, it's probably the best way to attract clueless but paying customers. Personally, I am vaguely nauseated whenever I engage in an IRC conversation; the atmosphere is dismally tacky, the people uninteresting, the conversations hideously dull and the software dreadful. (If IRC fans in the audience want to point out places where this is not true, they should feel free to drop me a line). After about a year or so of the above paragraph, a courageous soul has come out to defend IRC. Andy Church (achurch@dragonfire.net) mentions that the most prominent channels of IRC are indeed a mess, but there are a number of semi-hidden communities there where real conversation takes place. I had one of those myself, where the fans of my friend Jennifer's home page got together; that, too, was a lot of fun. So if you have a community of your own, and you want to talk together without running up massive phone bills, IRC does have its place. I'd recommend using the smaller but better-run Undernet for this purpose, however; that way you get more reliable connections. Or so I'm told, anyway. Despite all this - and probably because of it - the average user loves IRC, and it probably puts fewer demands on your system than just about any other service. As I would say, you don't have to like it to offer it. To add a little fairness to the above, I will say that IRC is no worse than any other multi-user chat system I've seen, such as those offered by various TBBS and Major BBS bulletin board systems. Still, commands such as "kick" and fearsome creatures called "bots" do not make for a pleasantly civilized atmosphere. IRC works by connecting you to an IRC server. There are two IRC networks: Eris-Free Net (EFNet) and Undernet. Because IRC is a networked resource that uses the entire Internet for its conversations, the two nets are the only points of difference; any IRC server on Undernet will be identical to any other IRC server there, and the same is true for EFNet. Choosing the closest server will not deprive your users of anything; it will, however, decrease loads on the Internet, and provide them with far superior performance. Because of this, taking a few minutes to find that server is strongly recommended before offering IRC to your users. I would suggest offering two IRC commands - one to hook up to EFNet and the other to attach to the Undernet. EFNet is the huge one; Undernet seems to be run by significantly more clueful people, but there's often nothing going on there. IRC was designed to perform operations enjoyed by the lowest common denominator, but since it is a typical Unix program designed by university types, it has tons of confusing options. I recommend that you study the Undernet's IRC FAQ, available at [[ pointer to undernet faq ] and make it easily available to IRC users. It should answer most of the questions that come up. Please note that the Undernet FAQ actually covers operation of the IRC system itself and not only Undernet. Even if you don't want to connect to Undernet (it's much less popular than the older Eris-Free net), you still want to read the FAQ and make it available. Fortunately, the Undernet people have also made it very easy to set up an IRC client. They have prepared IRC clients for various systems at their FTP site [[ find URL ]], and setup for them is clearly explained and quite easy. I recommend this strongly over trying to compile your own IRC client; I attempted that initially and ran into errors. The pre-compiled binaries from Undernet work just fine. As mentioned previously, IRC client software is not terribly well written and will very often "run away" from your users. This will cause it to fail completely, grinding up amazing amounts of CPU time in the process. See "Dealing with Runaway processes", below. +.# What about other chat services? With the increasing popularity of the World Wide Web all but eliminating people's knowledge of the older and less flashy Internet services, web chat systems have come to pass. Most of them are unspeakably horrible in implementation; nonetheless, they have gotten a quite respectable level of popularity. The newer generation web chat systems use Java, and if Java can actually be made to work well (the jury's still out in my opinion :-( ), this should be the chat of the future. A chat service called ICB (International CB) was created as a clone of the CompuServe CB simulator. It's a lot easier to understand than IRC; it's sacrificed features for ease of use. A good trade, in my opinion. Unfortunately, it seems to consist of the same vapid and boring conversations as any other chat system. You might find the idea of doing a local chat system a surprisingly good one; get members of your system to get to know each other, and build a community of sorts. This is especially interesting as a scheme because people who chat usually like to get together for meetings, something tough to arrange on the Internet as a whole. This is something I definitely plan to do on my system - if, of course, I can get some spare time to write the software! +.# Where can I find some interesting scripts and patches? Christopher X Candreva has put together a few perl scripts and patches that you might find useful. They include: dochk - Automatically processes checkgroups messages, and updates the newsgroups file. This is one of the most tedious and neglected aspects of being a news administrator. msgdesc - Update newsgroups file from one downloaded from another site, and/or to reflect the current active file. The newsgroups file includes descriptions of the newsgroups on your site. wwwpwd - Code to let users change their passwords from a web site. You can grab these utilities and more from the URL http://www.westnet.com/providers/ . + What about Fees, Terms and Conditions? I am shortly going to be writing a separate document on condition- related issues; hopefully I will be able to write a suggested terms and conditions list for all providers. +.# How much can I charge? How much do other providers charge? Most providers, including biggies Netcom and Earthlink, charge in the $ 19.95 a month range. Online services have fought back by also taking the $ 19.95 price point. I've seen ISP services for as little as $ 9.95 a month, but mostly for seriously limited access. Unlimited access is in the $ 12 to $ 30 a month range. The higher rates are charged by "boutique" providers that offer higher quality, more personalized service. Netcom has POPs in many major cities, so statistically it's the most likely competitor for you to come up against. Earthlink and CRL are other names you're likely to run into. When comparing your rates to Netcom's, keep in mind that they have had enormous service problems in the past, and will probably be continuously erratic. See elsewhere in this FAQ for details. New service providers seem to be charging in the range of $ 14-20/month for full-access Shell or BBS accounts. The main action in rates is that SLIP/PPP are now priced at within a few dollars of Shell access, with at least 20 free hours. Karl Denninger's MCSNet charges $ 60/3 months, or $ 240/year. If you stay with them for a year, you get two free months, reducing your effective cost to $ 17/month. For SLIP/PPP, the 14-month cost is $ 260, under the same terms. +.# How can I distinguish my service from the competition? In the beginning, low rates, friendliness of service and lack of heavy system load may be your best bet, at least when competing with the national providers. I'm using unique software I wrote myself as a lure, but I recognize that this is not an option for many people. (See the section on BBS software, below). +.# Unlimited versus Metered Access Before SLIP and PPP accounts became commonplace, most Internet providers had a single rate for unlimited access. During the days of the shell account, after all, if you were on the net, you generally couldn't use your computer for anything else. As a result, people would stay online for only as long as they needed. Because of this tradition, many providers have continued to offer "unlimited" SLIP/PPP access, with the important provison that this would not be the same as "full-time, dedicated, 24x7 access". Other providers have said that, because many people can camp online without even knowing they're doing it, some kind of time charge is necessary. There is a certain degree of acrimony on the Inet-Access mailing list between the two camps, with the metered access camp saying that the unmetered people are not able to provide high-quality service. This debate really matters little, as virtually all Internet providers are being forced to offer virtually unlimited access due to competition from other Internet providers and the commercial online services. One very important point is to be sure you define your terms correctly. An "unlimited" account has to be separate and distinct from a "permanent, dedicated" account. It must be made clear that you are not going to let people keep the the connction up 24x7 to run servers or MUDs. It's probably best to say that something like this: "There is no hard limit to the amount of time you can spend on your account. However, this account is /not/ to be used to run servers, and you must be on line for no more than half the day. We will not normally enforce that requirement unless your usage is continuously excessive for a long period of time." Richard Stiennon notes: "I have had customers that use IRC for *eight* hours a day. These "addicts" are usually good customers and bring in a lot of referrals. Heavy users can be a problem if your POP has only 20 or so modems, but once you are up to 100 modems statistics become your friend. Not a problem." David Graves (dgraves@netreach.net) comments: "One of the ways to insure that there isn't abuse is to call it "unlimited interactive usage" Which means that you will kick them out with 20 or so minutes inactivity. For the people who set their mail to read every 15 minutes to keep on -- well just set your term server to knock anyone off after 6 hours or so. If they're really on, they'll log right back in. If they're not there, then your modem is free for your other users. There's probably nothing you can do with people who have auto-relog on disconnect. "For us it just hasn't been an issue. "If you find someone staying on for 10 and 12 hours, you might email them and tell them that you really don't think that they're sitting there all that long, and would they please consider buying the 'dedicated access' package." +.# Accounting and Billing Practices (*) Billing by credit card has major advantages for both you and your customer. It's definitely the best option for most providers, but it may be exceedingly difficult to get. See below. There are a number of services now that will set up 900 number billing for you. Your customer calls the number and gets a code; he then calls your system and types the code in. The system is then unlocked for a specific period. You are paid at the end of the telco billing cycle. This is probably one of the easiest options to qualify for, but you only get 2/3 of the total charge; this might be unacceptable to you. People who have used the 900 number and received your services may also attempt to contest the charges on their bill. Due to the sleazy reputation of 900 number vendors and systems, this is normally granted as a matter of course by the local telco. Finally, you can get people to pay by check. You get 100% of the charge, and you get it now. The bad news is that many people who will blithely charge billions on a credit card have an amazing resistance to getting around to sending a check in the mail. Unless you have your strongman Guido around to collect from your customers, this might not be the best solution. You should always accept this form of payment, since it's the simplest type for you, but you should not count on your customers doing it. It's generally best to offer tempting discounts for longer terms, since it's a big hassle getting people to send in a check every month. +.# Credit Cards (*) There are two issues surrounding the world of credit cards. The first is how big a percentage they nick you, and the second is whether you can get them to accept your company at all. Most credit card processing companies are leery of any businesses selling services, especially out of the home. You will find it much easier to get their attention if you have physical business premises, instead of your garage. A company called Teleflora - that's right, a florist delivery service - has made quite a nice little sideline business out of processing credit card orders for bulletin boards and similar operations. However, they now appear to be out of this business (as of November 7, 1994). Ben Bradley is offering a service that appears to fill in the gap. He represents four "banks that specialize in granting credit card merchant accounts to ISPs and other on-line professionals. Makes no difference if the ISP is just starting business or operating out of their home. We can, in most cases, get them approved if they have good personal credit. We also offer Auto-Debit from the checking accounts of the customers of the ISP." For more information, drop a line to bradley@lords.com . Chris Hawkinson notes two interesting things. First, if you charge by the hour, you are counted as a worse risk than if you charge a flat-rate fee. To accept VISA, you must do the following if you have an hourly rate plan: * Pay a special annual fee * Have the worst discount rate * Have an 800 number which appears on credit card statements. Second, just because you have received money, it doesn't mean you won't have to give it back. Chargebacks can occur up to a year later with just cause. "Further, since an online system usually doesn't have the normal signature receipt, the customer is almost ALWAYS right." Finally, he notes that extremely tight security is necessary for a system that keeps credit cards information online, since crackers are attracted to card numbers like ants are to sugar. Perhaps the best way to do this is take the card numbers and copy them to a remote system inaccesible to the public. Chris Myer (cmyer@su102a.ess.harris.com) did a complete survey of net.wisdom on this topic. He came up with the following additional comments on how you might get accepted for a merchant account, and what kinds of rates: 1 Get an account through a local bank. Chris says this isn't very likely, unless you have very substantial dealings with them. "These guys don't understand the 'net, and will simply consider you a mail-order business. If you are home-based, forget it. Your only hope is to get an employee of the local bank to go to bat for you--in a big way!" Karl Denninger apparently got his own merchant account in this manner; keep in mind, however, that his five-digit account balances probably helped just a little. 2 Independent Service Organization. These people serve as go-betweens, insulating you from the more stringent standards banks require. According to Chris, "Many of these companies charge horrendous up-front, non-refundable fees, but some are reasonable." The most well-known of them is Teleflora. Evidently these guys decided that they would become ISO's, and they are big believers in BBS's. Fairly large non-refundable application fee. Unfortunately, they seem to be leaving the business at this time. (800) 325-4849. (x2076?) Card Service is another outfit that's been mentioned. Unfortunately, little information is available about their serivces. Card Service, (800) 944-7164, or try (800) 947-3650 (faxback, request 765.) 3 Discover Card: "Evidently, they not only provide Discover cards, but once you get one of those they will also provide access to Visa/MC merchant accounts. Call (800) DISCOVER (347-2683 for the alpha-numerically challenged) and ask for Merchant Services." Here's the information from Chris on what kind of rates you can expect to pay: Samples of Rates Charged: ========================= Visa/MC: 2.20% 2.25% plus $.25 per transaction, plus $2.00 statement charge. 2.32% plus $5.00 per month. 3.0% 3.5% plus $.30 per transaction. 4.0% Discover: 1.9% 3.25% AMEX: 3.25% 4.5% Terminal: $21 per month, or $200 outright. $600 outright, $75 for repair. The following is copied straight from Chris' message: Suggestions for Getting Accepted: 1 Try to avoid being labeled "home-business", "mail order" or "telemarketer". I don't recommend lying if you fall squarely into one of these catagories, but don't let them put you there if you're not. 2 Minimize the amount you predict you will charge. Due to the fact that charge-backs can be made for up to 6 months after a transaction, the bank's amount "at risk" is 6 times the predicted monthly amount. 3 When working with a bank be prepared to keep liquid assets in the bank equal to or greater than their "at risk" amount. This is usually necessary for no more than 6-12 months of good charges. Suggestions for Protecting Yourself: 1 Try to find some way to get a signature on file. One way I thought of was not only to have them fax in the card number and their signature, but a photocopy of their picture ID with signa- ture on it. 2 Make sure you have a way to verify that the ship-to address for your product (if you are a retailer) is the same as the billing address of the credit card holder. +.# Billing Software A more up to date list of billing softare is at http://www.isp-invoicing.com . There are a number of billing programs out there for Internet providers. A company called Interstate has written a "thorough billing system called 'im', or 'INC Manager'." It keeps track of billing, pro-rating, account creation, expiration and bill and label printing. Cost is $ 125/site license. [I have gotten several enquiries for company contact information; unfortunately, all I know is what I wrote here. Perhaps someone from Interstate can come forward with something?] A system called BATS (formerly IOU) is promising, but still under development with some serious bugs, according to a beta tester of the system. You can learn more about IOU http://www.astroarch.com . Edward L Haletky , speaking for IOU's developers, AstroArch Consulting, says, "Any problems currently referred to in the FAQ have not existed for a very long period of time." IOU supports BSDI 1.1 and 2.0, FreeBSD v2.0.5, AIX v4.0, HUPX v9.0.x, IRIX 5.x, Linux, MachTen, OSF/1 v3.2, SCO, Solaris 2.4 and Solaris x86, SunOS 4.1.3, Unixware, Digital Unix and MachTen. The current version is 2.0. I'd appreciate comments from other users so I can provide an impartial update of IOU's status. So far, nobody seems to have been upset enough to give me additional information, so I'll assume the software has overcome its previous problems. The name IOU has now been changed to BATS, "for various reasons". User Tracking and Accounting from RTD has had some problems with support, so they've thrown it into the public domain. It's a bunch of perl scripts that runs under Unix, and includes full source code. If you're a hacker, I'd definitely recommend that approach over using a Windows-based program. Check out http://www.rtd.com/software/uta.html Keven Fink has just announced the ISP billing system written in Perl. Here's the summary he sent us: "N2H2's User Registration, Information, and Billing System (URIBS) is a complete user administration package for UNIX-based ISPs. It allows non-technical staff to create and administer both shell and dial-up user accounts. It also provides billing and accounting capabilities. "The package is written entirely in Perl, so all source code is included to allow easy customization. It has been tested on a variety of UNIX variants, including SunOS, Solaris, Linux, FreeBSD, and BSDI. It was designed to work with RADIUS for dial-up authentication and accounting." For more information, check out http://www.n2h2.com/URIBS/ . There is another system called IAF by Solect; more information can be found at http://www.solect.com/ Imagen - http://www.imagen.net/ has a new billing system. Once again, it's Windows-based, alas. However, it's worth noting that it will read industry-standard RADIUS and web log files, which means that you don't have to use NT servers to run the program, despite what its web page might imply; instead, you can go to your Windows (95 or NT) box, give it the files from your Unix servers, and it will grind away on them. Since it's a Windows program, source code is hidden and you cannot fool with it yourself (unlike the Perl systems mentioned above). They use an Access database, so it's a bit of a slug, but you can use standard query software built into Access to build your own reports. Note that, unlike Coolworld, the database is open and can be played with at will. Access is not required to run the product. However, it might be a good choice for non-hackers; I did get the impression that the people who suggested it to me were responsive. So if you must use Windows for this crucial function, this might not be a bad choice. One special feature it has that's worthy of note is that you can use it to bill by the hour for applications like cyber cafes and such. The system called Coolworld - http://www.coolworld.com - has been criticised heavily by its users on the Inet-Access mailing list. It's Windows and Access-based, which in my opinion should disqualify it from consideration by any serious ISP. Those who disagree with this analysis should note that the database files are apparently locked, so you cannot alter or modify the files using any kind of outside program. Coolworld users found that it was impossible to import historical information from other packages, and that technical support was unhelpful. Coolworld has threatened to sue a person who posted what they considered to be derogatory information on their package to the Inet-Access mailing list. I would recommend in the strongest possible terms that you stay away from this company. Intbill is a package originally suggested by Emil Mikhles . Its URL is http://www.coolworld.com/intbill/ . It, however, runs on Windows 3.1, Windows 95 or Windows NT, so Windowsphobic people (like me!) may not care for it. It requires 16MB of RAM. Emil reports that he no longer recommends Intbill; he now uses a custom package written in Access, linking the data to PC-Charge (a Windows equivalent to IC Verify). Kevin Fink of n2h2 (see above) has created his own list of every ISP billing program on the market: http://www.n2h2.com/URIBS/others.html . On 12 September 1997, James B Hrdy posted the following list of billing packages to inet-access, unfortunately without any commentary. This should give you some more help with your quest: http://www.cyberacs.com/ ISP Trak http://www.xpert.com/ X Acct http://www.digidyne.ca/iapplications.html Tips http://www.cybertimesoftware.com Cybertimes http://n2h2.com/URIBS// Uribs http://www.mcmsoft.com/isovr.html MCM REAL TIME Accounting Systems http://www.interbiller.com/ InterBiller http://www.pacnet.com/pacnet/wcube/screens.html WebCube Infinite Patience You are going to be dealing with unreliable software being used by largely clueless Microsoft Windows users. You will need to give high quality phone support to these people, or they will never manage to figure out how things work. This is probably the most severe problem with offering SLIP > A starter kit With the advent of Windows95, it's gotten a lot easier to hook up to the Internet. Windows95 includes a functional TCP/IP stack, functional (albiet not graphical) ftp, telnet and traceroute (tracert) commands, and a fairly straightforward setup. For the setup that even an elementary school kid could use, just add the Plus Pack, which also includes an obsolete copy of their Internet Explorer browser. Because of this, your Internet starter kit would appear to be far less important than it was in the past. However, competitive pressure has appeared since most large providers offer a licensed copy of Netscape with their service. Smaller ones, however, find it difficult to afford Netscape because you can only get good price breaks on it in quantities around 1,000 or more. Inking a five-figure contract with Netscape is not most small ISP's idea of fun. Michael Dillon has an update on this; apparently the dust has cleared and you can now get quite decent terms from Netscape resellers such as Intercon. "I have in front of me a price sheet from Intercon that includes Netscape, Eudora, ftp client, news client etc... for $14.95 in quantities of 100 and up. Even the smallest ISP should be able to commit to buying 100 units and $14.95 is low enough that you should be able to easily pass this cost on to customers as an account setup fee. And it is available for Windows and Mac. Send email to randall@intercon.com for details and maybe he will send you the demo kit that was being handed out at ISPCON." This seems to me like an excellent idea. One of the big questions floating around the Internet provider world, then, is whether you make a deal with the devil. Microsoft will give you a free license to distribute Internet Explorer, their browser, and all they ask is that you add a smattering of their extensions to your home page, stick on a "Best Viewed with Internet Explorer" label, and not recommend any other browsers to your customers. Those who are not virulent Microsoft haters have taken a hard look at this package, although there don't seem to be many defections from Netscape; it's mainly people who didn't have a browser deal before. In truth, Netscape has been remarkably insensitive to the needs of the small ISP market, and this may hurt them badly in the long run. MSIE is now said to be quite a decent program; it actually pioneered some very useful extensions, such as <table background> and popping up the document text before the background loaded. Netscape has since copied these, and since then it's been a wild battle to the death between the two companies. I have very mixed feelings about all this. I don't like what Netscape's done to the small ISPs, but I don't trust Microsoft any further than I could throw them. If it was my call, I'd provide no software and have my customers FTP Netscape. The problem is that Netscape is very, very slow to download, so this can be a very annoying option for many people, especially those just starting out. This dilemma may well be resolved, but not in a way I'll necessarily like: Soon, Internet Explorer will be included within every release of Windows95. We'll like that, because we'll finally be liberated from the need to wrestle with troublesome installation disks. But it may well mean the final victory of Microsoft in the browser wars -- strictly by default. Unfortunately, Windows 3.1 customers are probably still the majority, due to the unfortunate reputation of Windows95 in many quarters. As a result, a Windows 3.1 setup kit is still needed, and Microsoft has attempted to provide one. Early reports, however, state that the TCP stack is riddled with problems, which might well make you think twice about signing that Microsoft deal. So support for Windows 3.x is still probematical, and the following older section of this document describes what's usually done. SLIP/PPP software is complex. In order to offer SLIP, you have to supply your users with a disk or disks containing software to install on their computers. (Since I first wrote this, PPP has largely replaced SLIP, mainly due to its inclusion in Windows95). Most of this software is shareware, and there are some complex distribution conditions. For the most part, you cannot distribute SLIP software pre-configured for people's systems without paying a license fee to the software's owner. Some of the software, including Netscape, cannot be legally distributed on a disk without a license - and licenses for Netscape are expensive: $ 17/copy with a minimum buy of 500 copies. What most people do is distribute a rather minimalist startup kit, normally including Trumpet Winsock (said to be the least bad of the available TCP stacks) and public domain FTP, Gopher, mail and WWW clients. Customers are then given instructions to set up the disk, and can then download Netscape from the main Netscape www site. Note that the version of Netscape sold on a retail level has startup access for various commercial Internet providers such as Netcom and CompuServe; its configuration procedure is apparently severely biased against local ISPs. This might be worth keeping in mind when deciding whether to use those Netscape extensions ... Microsoft is providing a free startup disk kit called the IETK. People who have used it seem to be having almost constant problems setting it up and trying to get it functional. It's free, but you have to use at least one of Microsoft's IE extensions on your web page and state that your site is bets viewed with MSIE. On balance, I'd rather go with Netscape despite the cost. Due to the ever-increasing size of this software, it's increasingly cost-effective to burn startup CD ROMs instead of disks. The IETK, for example, takes five disks. At a cost of $ 0.35 per disk, that's $ 1.75. Pressing a CD ROM is not only less trouble for the user, it's also a lot cheaper; you can do it for less than $ 1 per CD. +.# Do any companies offer starter kits, or do I have to roll my own? Soorena Salari offers InetMgr, an installer package that makes the connection easy. It creates the proper directories, installs the files, and asks the needed questions. Cost is $ 2/disk, quantity 1,000+. A test drive version is available for your trial at http://www.ccsweb.com/inetmgr . Emil Mikhles recommends Better Internet Starter Kits at http://www.talon.net/iqcdemo . +.# How has Windows95 changed the SLIP/PPP picture? Windows95 includes dial-up networking out of the box. As a result, configuring new PPP users (Windows95 doesn't use SLIP) is quite a bit less difficult than it's been in the past. A site called Windows95.com [ http://www.windows95.com/ ] has an excellent set of instructions which I believe you can provide to your customers. They'll save you hours of hideously boring technical support. The Windows95 setup wizard included with the Plus Pack makes setup a nearly brainless experience. All you need to do is type in the IP numbers of your various servers and you're all set. You can talk new subscribers through this over the phone in about five minutes. If you don't get the Plus Pack, the instructions on http://www.windows95.com/ are worth their weight in gold; you can still do it, but it requires wandering through a whole bunch of tabbed dialogue boxes. If your users are currently using Windows 3.1 and upgrade to Windows95, they will find that their existing Windows software almost certainly will not work. To fix this, they can either set up Windows95 dial-up networking, or find their old winsock.dll file (which I believe is renamed to winsock.old) and overwrite the Windows95 driver with it. If your users presently have Trumpet, I strongly recommend Windows95. Its dial-up networking is significantly easier to set up than Trumpet, and it seems quite a bit more reliable as well. +.# What about proprietary GUIs, such as the Pipeline? The Pipeline, a NYC-based Internet provider, took a different approach. They wrote their own customized terminal program that basically makes their system look like SLIP/PPP without all the headaches. It's clever and works well. Unfortunately, as of last writing, they were asking a rather massive price for it. You can try it free of charge on their demo line; telnet to pipeline.com. At one time, I was planning to write a similar GUI front end for my own customized software. However, the stampede towards Netscape and PPP connections has made this look like a fairly poor idea at this time. The Pipeline was sold to PSI and is now, alas, out of business. +.# What about BBS Interfaces? I wrote and ran a customized BBS between March 1994 and June 1996. Unfortunately, the market for the type of system I had pretty much vanished after people started selling SLIP/PPP connections for $ 9.95 a month. Because of that, and due to a hardware failure that made the system uneconomic to keep up, the system is now down. Shed a tear; it's been completely replaced by my Web provider. A spiritual successor to my BBS can be found in my web personals pages, at http://cgi.amazing.com/personals/ . The rest of the information in this section is from 1994 and is probably very much out of date. The BBS world as a whole seems to be dying with the dominance of the Internet. For those who still care, esoft, makers of TBBS, now no longer sells or supports the product, so we can't recommend it anymore even if you just want to run a BBS. Pity since it had the easiest and best customizability of any commercial BBS software. Bulletin board systems, such as TBBS and Major BBS, are rushing into the Internet bandwagon. Traditionally, the interfaces used by both these programs are far too clunky for efficient newsreading, and that is virtually the only service they've been able to offer. How much this changes will determine how much influence BBS vendors finally have on the Internet world. Eric S Raymond has been working on some similar projects. Like many projects of its kind, this was inspired by dissatisfaction with existing systems: "I know of three major projects to produce a UNIX BBS interface suitable for use by ISPs -- FreePort, InterLink, and Chebucto Suite. "FreePort is the BBS interface used by NPTN, the National Public Telecomputing Network, and its affiliated Free-Nets. Despite its name, it costs $850 to license from Case Western Reserve University. And it is an utterly awful implementation of a clumsy, stupid design. Most of the traffic on its lists, freenet-tech and freenet-admin, is disgruntled Free-Netters wishing desperately they had something better. To see for yourself, rlogin as `visitor' at freenet.fsu.edu or yfn2.ysu.edu." I - the FAQ maintainer - wanted to check out this claim, so I did exactly as Eric suggested. Sure enough, the Free-Net software is awfully clunky for an $ 895 program. It's inarguably simple, but the user interface is far from the best. It looks a lot like a stripped down version of Galacticomm's Major BBS; unfortunately, that's not a compliment. Eric again: "InterLink is the interface I wrote out of complete disgust with FreePort. It combines the ease of use of a menu-based system with the feel of a MUD (multi-user dungeon) game; it has unique features for decentralized administration; and the code is simple, flexible, and easily extensible. This software is production-quality, supporting more than 1000 users as of November 1994, and is far superior to FreePort in every respect, but I have not yet released it; I'm in the midst of a redesign for better integration with WWW. To check it out, telnet to locke.ccil.org." I checked it out, and it is indeed a slick, nicely designed and easy to understand program. It has a much more modern look and feel than the FreeNet software, and it's considerably easier to figure out. The David's Amazing Internet Services software - otherwise known as my own - is very different from either program. Anyone interested in checking out the state of the Internet BBS art should definitely take a look at all three systems. Which one is better? I'd say it depends on your own personal taste. Eric continues: "There is a project called `Chebucto Suite' under development in Canada, with goals similar to those of InterLink. From their prospectus, it appears that they already have full WWW integration (the Chebucto browser is a souped-up version of the WWW lynx client). They appear to be behind InterLink in some other areas, notably administration tools. I don't know where Chebucto Suite is actually running; contact David Trueman for information. Earlier, Eric was quite keen on integrating his project with Chebucto Suite. However, he recently wrote the following: "The Chebucto people seem to have fallen into a hole. They're not answering mail, and I've heard nothing about them on any of the FreeNet lists." Lasse Morkhagen tells us that BBBS is a European BBS program from Finland, which supports News, Email, telnet, rlogin, flexible accounting, Fido-Net and a C-like scripting language for customization. To try it, you can telnet to fix.no . +.# What about menu interfaces? Many people are now using a World Wide Web browser, such as Lynx, as their main menuing system. This is easy for most people to use, and can take them to all sorts of interesting places. Some people have trouble with this because Lynx won't work properly unless your VT-100 emulation is exactly perfect. However, the flexibility and ease of use are certainly there. A company called Interstate has written a menu shell in Perl that features complete full-screen operation. A site license for it costs $ 75. Unfortunately, I lost the message with further information. +.# Is there an easy, painless way to provide Internet services? Possibly. Draper Kauffman has a beguilingly simple business idea to share with potential providers in underserved regions of the world: sell telnet-only accounts to his system. This way, he notes, you lose a lot of headaches. No need for gigabyte drives; use his. No need to offer shell accounts and stay up all night worrying about infamous security holes; he'll offer them for you. No need to contract with credit card companies; he knows how tiresome that is to new businesses and takes all of it out of your hands. You charge what you want and it all goes through his accounting system. Anything over a $ 15/month flat fee is given to you as a nice convenient check you can deposit right into your account. The main advantage of this idea is that you have 24 hour a day 7 day a week monitoring and support by his already-existant ops center. This would be horribly expensive for you to do yourself. The main disadvantage of this idea is that you have no control over your own system; in reality, you're selling someone else's product and not your own. If you (like me) are starting a provider to put your own personal stamp on the world, this is definitely not for you. However, if you want money and you're in a market with low competition, you could probably charge $ 25/month and pocket the $ 10 difference without doing anything more strenuous than maintaining your modems and router. Note that, since this would be a telnet only situation, you would almost certainly require a T1 connection, since 100% of your users' time would be using the Internet itself quite intensely. The high bandwidth required might offset some of the administrative savings of the proposed plan. However, it can be an effective way to build a strong customer base at the start, where cash for hardware, software and customer support staff is hard to come by. You also have more time for marketing your service, and it's easier to sell a full-service account for $ 25 than a semi-reliable account on a start-up system for $ 18. Draper suggests that your initial goal should be to build up your system in stages to handle 750 to 1,000 customers using 64 lines and modems and a T-1 line. If you charge $ 25 per customer and pay about $ 30/line and $ 2,000 for your T-1, you net somewhere around $ 4,500 to $ 6,000 per month with no payroll. Then, with this kind of strong cash flow, you can easily add the drives, CPUs and staff you need for complete services, and gradually bring your customers back to your system full-time. A Phoenix, AZ company called Primenet is now selling Internet franchises, which work in a very similar way to Draper's plan. I have seen their franchise agreement, and it appears to be incredibly slanted against the franchisee. I urge anyone interested in that franchise (or any other) to read the agreement carefully before deciding to invest. Another company called Internet Online Services (IOS) has also been forming cooperative agreements with users; unfortunately, their very low rates and massive national advertising has resulted in horribly poor service (see http://www.thelist.com and look them up for the gory details). Draper Kauffman himself seems to have disappeared, and I have heard from some io.com users that service has deteriorated significantly in recent months. + What sorts of technical problems should I expect? Thanks to Tony Sanders for contributing this list of some of the many problems Internet connectivity is heir to: Internet connection line failures (dealing with telco and provider) Routing problems General network problems (site x.y.z is down and the user complains to you) Catastrophic Disk failure (make sure you have backups) Users deleting files and then wanting them back (backups again) Modem problems all of sorts (parity mismatch, hung modems, flow control, not disconnecting properly, not connecting properly) System crashes and hangs Configuring ftp, telnet, shell accounts, nntp, www, smtp/sendmail, nfs Bug tracking (a potentially big problem) Billing +.# What can be done about System Crashes? Ideally, there should be someone around 24 hours a day to make sure the system is put back up when it dies. A reasonable compromise for companies that aren't large enough to do this is to be sure someone's on the system during peak usage hours, to make sure the system is reset when it goes down. (Information on devices to reset the system automatically upon a crash should be given here). You may be able to prevent many system crashes - or at least minimize their effect - by purchasing an Uninterruptable Power Supply (UPS) and a mechanism to automatically shut the system down gracefully upon power failure. Unfortunately, this is another issue I'm not that familiar with; expertise from those who know would be appreciated. Walter Vose Jeffries has an interesting suggestion that I'm surprised UPS companies don't try packaging. He writes: "What we did with our systems here is bought cheap ($ 500) but reasonably good UPSs and then replaced their batteries with deep cycle marine/truck batteries. This gives us 8 to 16 horus of run time on the battery (which then takes 4 days to recharge completely). This works well and is cheap; the batteries cost $ 60 to $ 100 or so. That's much less than the cost of a big UPS and by having several that each give long power protection we are even better off than with one central unit. In seven years, our longest outage was 7 hours with three others at about 4 hours and most 5 minutes to one hour. This way, we don't worry about having to shut down since we know we'll outlast even the worst power outages we've had in seven years. (Of course we still program in graceful shut down - but better to never have it happen)." I would have loved to have had such a thing during our 1994 earthquake, where power was out in many places for as much as a day or two. Judicious manipulation of multiple deep cycle batteries probably would have kept my system going after switching off all non-essential equipment. For example, if I'd had a news machine at the time, I could have deactivated it and used its separate UPS when the main system was almost out of power. Walter tells me that you can also charge batteries during long outages by just sticking them in a car, or you can start your car by "borrowing" one of the UPS batteries! During my current business trip, when I've had to be away from the system for almost two weeks, I asked my roommate's girlfriend to check the system and reset it when "nothing appears when you hit the key on the console." Apparently, she's already done this once, and it's worked fine. If you can explain the care of your system easily enough, just about anyone who likes you should be able to help. (I wouldn't recommend handing this off to an enemy, since it's all too easy to destroy your system this way! :-( ) +.# What can be done about Network Outages? When part of your network link fails, you're dead in the water. If you're using a SLIP connection for your network link, you will find yourself disconnected occasionally. I have written a program to automatically reconnect myself when this happens. It runs every 15 minutes (through cron) and checks to see if there is a DIP process running. If there is no DIP running, it starts one up. This lets my system automatically reconnect even when I'm not there to tend it. Here's the program: /* dipcheck.c -- check to see if DIP is running By David H Dennis * david@amazing.amazing.com This program is hereby placed in the public domain; no warranty exists, expressed or implied. */ #include #include #define FN "/tmp/dipps" #define LEN 100 main() { char s[LEN]; int ct = 0; FILE *fp; system("ps -aux | grep dip >/tmp/dipps"); fp = fopen(FN, "r"); while (fgets(s, LEN, fp)) { ct++; } fclose(fp); /* The PS and DIPCHECK commands also contain the word 'dip', so if there are less than 3 uses of the word found by grep, we're not connected, and an attempt should be made to reattach ourselves */ if (ct < 3) { printf("Executing DIP ...\n"); system("/user/dip/dip /user/dip/sample.dip >/tmp/dipout"); /* Note: Hollywood is the name of my network connection */ system("route add hollywood"); system("route add default gw hollywood"); } } Someone will probably flame me for writing this in C, when it would have been more elegant as perl or even a shell script, but who has time to learn them? :-( According to Tony Sanders , "Perl is a godsend for ISP's willing to learn it. It is the swiss army knife of the system administrator, a real survival tool." I finally broke down and learned it to write the "kill runaway processes" program, which will be presented later. If a 56k or T1 connect fails, Tony Sanders says that you first reset everything on your end. The CSU/DSU (at least the ones I've seen, cheap models might not have it) has various local and remote loopback tests it can perform, do those. If those fail you call your provider and/or the telco. +.# Hung Modems Rackmount modems really score here; as mentioned previously, there are complex diagnostics and re-routing systems built into the modems. If, as seems more likely, you have a bunch of tangled wires leading to heaps of external modems, you will have to find out which one is causing the problem and reset it. This can usually be done easily enough by switching it off and on. If you are calling the system from a remote site, and find it rings and rings thanks to a bum modem, you can transfer to the next line by calling the main number on your voice line, and then calling the same number on your data line. You should then get the next line on the rotary, which is (hopefully) active. Then, it's a fairly simple matter to inspect ps, find the runaway job, and kill it. Usually that will reset the modem, and the system will once again work. (Again, Walter Vose Jeffries notes that the Call Forward on No Answer solution would work even better, since users would never be inconvenienced). If you still can't get on, it's recommended that you dial up a backup account you have on a competing provider and telnet to your system. I have such an account on both Netcom and Smartdocs (the latter being a small local provider). This also helps me test customer complaints about reachability, and problems I may have with my WWW pages and other services. +.# Killing Runaway Processes As we discussed in the sections on IRC or Lynx, these programs have some interesting bugs that cause them to "run away", making CPU usage zoom to no great use. I have devised a Perl script to scan PS in search of these evil jobs. It consists of two parts: RCHECK runs RUNCHECK repeatedly. Because I am still testing and refining the performance of these programs, I run rcheck from a virtual console, and occasionally watch its work. Once you're satisfied with it, you can put rcheck's single pipeline (the first 'system' command) in your crontab and run it every 15 minutes or so. Note that this has two separate code segments, (A) and (B). (A) kills any process that exceeds the CPU time listed that is not being used by administrative users (remember to put your own name on the list!). (B) kills any irc or lynx processes that are not being used by an administrative account. (B) is recommended if you sell shell accounts. NOTE! You must select one and only one of (A) or (B) - comment out the other by putting "#"s in front of its lines. You may want to run runcheck with the killing parts commented out to see what tasks it actually kills before using it. These are my first perl programs, so be gentle with criticism. In particular, I'm sure rcheck could have been written better without the system commands. As always, these programs are freely given to the public domain, although it would be nice if you kept the credit lines in. Since I didn't sell these programs for billions of dollars, of course I accept no responsibility for the consequences of trying 'em out. rcheck: #/usr/bin/perl # rcheck - run runcheck perl script forever # By David H Dennis ; # from the inet-access FAQ while (1) { system("ps -aux | perl runcheck"); system("sleep 15m"); } runcheck: #/usr/bin/perl # Perl program to process output of PS # By David H Dennis ; # from the inet-access FAQ system("uptime"); system("date"); while () { chop; @ps = split; if ($ps[2] > 15) { print($_, "\n"); print("Excessive CPU usage - candidate for killing\n"); if (($ps[0] eq "news") || ($ps[0] eq "root")) { print("... but news or root process, so ok\n"); } # You can exempt other commands in a similar way ... elsif ($ps[10] eq "bbs") { print("... but BBS process, so ok\n"); # and users as well } elsif ($ps[0] eq "david") { print("... but David's process, so ok\n"); } # Pick either (A) or (B), not both # (A) Kill any high-CPU process - use only if you run BBS accounts only else { $x = $ps[1]; print("kill -9 $x\n"); system("kill -9 $x\n"); } # End (A) # (B) Kill only IRC or Lynx processes - use if you sell shell accounts # Note: This has not been tested and is for your information only elsif ($ps[10] eq "lynx" || $ps[10] eq "irc") { $x = $ps[1]; print("kill -9 $x\n"); system("kill -9 $x\n"); } else { print("Process not killed - not in list of killable commands\n"); } # End (B) } } +.# The Dreaded Disk Space Crunch Even mighty Netcom, with Reiger-knows-how-many gigabytes of disk, has run out of space on occasion. So it's not just you. (Bob Reiger is the owner of Netcom). Don't let that make you feel complacent, though. There's little you can do to your users to make them more unhappy about you and your system. Bryant Durrell writes us as follows: "When you enable FTP, you too will discover the joys of people downloading the entire Linux distribution. Urk! Disk space is always going to be a problem, whether for this reason or another. You'll need to keep an eye on it, and decide what to do when it gets low. (Users will want you to buy more disk space)." I might add that you'd better check up on your i-nodes as well as your overall disk space. Because I didn't, I've lost mail on my system. Don't let that happen to you! +.# DNS Blues Domain Name Service (DNS) software runs invisibly on your machine and normally gives little trouble once you have it set up. It is recommended, however, that you ditch your vendor-provided copy of BIND (the DNS software) and go with the latest versions, which can be found somewhere on ftp.vix.com . The InterNIC, however, is another story. Some people have received excellent service from the NIC, while others report unanswered mail, unreturned phone calls and unseely delays. In the last six months or so, the InterNIC seems to have cleaned up its act pretty well. Now most domain registrations get accepted within 24 hours, a tremendous improvement over past NIC behaviour. There have been significant discussions of adding new Top Level Domains (TLDs), such as .bus, ,biz, .web, etc. An organization called the AlterNIC (http://www.alternic.net) has been formed to support additional TLDs. Unfortunately, this rather chaotic effort isn't connected to the standard root name servers, so users have to connect to the domains by either using special root server tables, or by typing in the IP addresses manually. So far, this service has met with big yawns from those who manage such things, so I don't recommend you do business with them. It has been widely rumoured that a new policy on additional TLDs will be announced, and hopefully at that time there will be room for competing registries. +.# What would be a good backup policy Off the top of my head (another section that needs to be fleshed out with some real-world opinions), you should back up your system and user files daily, probably with a seven-day rotating backup procedure. I wouldn't be worried about news; lost news tends to be a self-correcting problem. Recommendations on backup equipment and procedures would be appreciated here. +.# What services are particulary hard on performance? What should I do about them? My thanks to Alicia Salomon for reminding me to ask this question, and supplying part of the answer. Most providers will start with a single computer performing all functions, including mail, news, ftp, www serving and user processing. Because news flows into the system constantly, and since its processing can put a significant burden on the system's disks, this is normally one of the first functions to be transferred to a separate machine. Since the advent of INN, this is not nearly as much of a problem as it once was, but this is still sound advice. Tony Sanders notes that news is *very* hard on your network link unless you have T1. It's about 25% of a 56K line. Tony recommends PageSat, as shown elsewhere in the FAQ. PageSat, however, has many dissatisfied customers. The most common reference to PageSat is that service is poor, but the service definitely does save bandwidth. Rain obscures the signal, so on rainy days or during other unfavorable times, you'll still need to get news through the Internet. The bottom line is that it's very worthwhile on a 56k, because bandwidth is at a premium. You're not likely to need or want it if you have a T1. For an impartial view of PageSat, check out Nick's PageSat web page at http://www.kfu.com/~nsayer/pagesat/ . This page was created by a PageSat user and clearly explains the pros and cons of the package. My thanks to annette@acm.org (Annette Thompson) for pointing this out. An update on PageSat was provided by Kevin Kadow . "While it used to be a good deal at $ 1,000 setup and $ 300/year for a full feed, they only have a 9600 baud channel, so even with compression it's been falling behind. Now the company wants to switch to a 115kbps connection, forcing all the customers to buy a new $ 800+ receiver and probably a terminal server as well, since few PC or workstation serial ports can keep up with a compressed 115k stream!" Tragically, PageSat now appears to be out of business. I guess we can call this one an excellent idea gone sour. Other processes often put on dedicated machines include FTP, Gopher and WWW. FTP in particular can put an enormous strain on system disks, especially if users are allowed to place popular files in their own directories. There was an enormous stink created on Netcom when it was discovered that some users' FTP directories had X-rated pictures in them, and that they constituted some 60% of the total bytes downloaded from all of Netcom, causing vast overloads on Netcom's machines. It might be a very good idea to devote a machine with a large local disk to the user directories and transfer all the load off the main system. Unfortunately, this doesn't help ease the strain on your net connection. Potential load from Gopher and WWW could be immense, particulary if massive image files are involved. It would probably be a good idea to use the FTP machine for user Web and Gopher pages as well. Tony Sanders notes the following: "Well, the real point of load comes from how popular the information is. The servers of the Shoemaker-Levey comet photos got creamed as thousands and thousands of people requested the pictures. The good news is that you can charge some serious money to local business to put up information on WWW. That alone could probably pay for a T1 line." Multi-User Dungeons, or MUDs, are "virtual world" games that account for a large percentage of the Internet's popularity. Karl Denninger allows them on his system because he had a spare SPARC 10 sitting around with nothing better to do. If you are not so lucky, MUDs could devour your system memory and effectively kill performance totally. Here is another case where a dedicated machine would seem almost a necessity. Note that telnetting TO MUDs is a very easy thing to do; operating one on your system is the complex and compute-intensive burden Karl's taking about here. --stopping point-- +.# What sort of hardware should I use for my news system? This section is now laughably out of date. People who are serious about news are now using 30 gigabyte or more RAID arrays. Yikes. Karl Denninger gives a frightning vision: He uses a P90 with 64MB RAM and around 10 gigabytes of storage. Some people on the Inet-Access mailing list now (August 1996) go further; many run whole SUN SPARC 20s with multi-gigabyte RAID systems. Most of us, of course, would like nothing more than to follow that example; the only problem is that our checking accounts are suspiciously bare. Craig Warner notes that Clark net, an Internet provider, uses a SS20 with 192MB of RAM and a massive amount of disk. Most of us don't have that kind of money, either. The key to having a cut-rate news server is short expiration times. Craig tells me a SS1 with 32MB RAM and 2-2GB drives would do fine, if you expire your news in about 2-3 days and have a reasonable number of readers. If you have about 5 or so readers (which would be appropriate for up to 20 lines, most likely), you could get away with that system and a 14 day expiration time (although you might need more disk space than that). Once you get more readers, you need a more powerful news machine. The best idea by far, though, is to run news off your provider's machine or someone else's. News saps a tremendous amount of your bandwidth, and most newer net users are interested primarily in surfing the web. So having your users individually read news from your upstream provider's server would be a better way to get started - and you could be spared an enormous administrative burden. If your provider doesn't want to do this for you, there are at least two companies that will provide this service: alt.net : Contact Chris Caputo (ccaputo@alt.net) for information. (This service has been closed to new users for some time). texas.net: Contact barron@texas.net (Jonah Yokubaitis). $ 50/month, $ 0.20/user. dgs.dgsys.com: Justin Newton , (703) 749-2884., $ 50 flat rate for small ISPs. inquo.net: Contact info@inquo.net. News hosting: $ 100/month + $ 1/user. Full (10k+newsgroups) newsfeed or hosting on a *very* fast server with a T1 line for high speed news access. A full newsfeed is also offered for $ 70/month. dbtech.net: Contact dbrass@dbtech.net. News hosting $ 40/month + $ 0.20/user, $ 0.44/user for Clarinet. Carries alt, rec, talk and sci groups minus "questionable" ones [in Alabama] including alt.binaries and probably most of the sex stuff. This might be good for people who are strongly anti-porn or believe the Exon stuff is the wave of the future. Note, however, that no newsfeed service can guarantee fully filtered content. +.# What can be done about users who spam? See my separate document, http://amazing.freelink.net/internet/spam.html . +.# What can be done about users who walk away from the keyboard? Write or obtain an idle timeout program. Usually the archives for your operating system will have something that will do. For Linux, ftp to Sunsite.unc.edu and get /pub/Linux/system/Admin/idleout.tar.Z. There is a certain degree of controversy over how long the idle timeout should be. Netcom uses 10 minutes, which many people find too short. MCS uses 20 minutes, which is probably about right. I think it would be a good idea to vary the idle timeout depending on the number of lines in use. During an extremely light load time, it might be ok to make it as much as an hour. This can help users who have to go to the bathroom or who got engaged in a long conversation, and it doesn't much hurt the system. However, I have not yet experimented with the idle timeout software. Tony Sanders agrees with this idea, and suggests that you might want to involve users in the process. Explain to them why an idle timeout is necessary, and get their input. It is, after all, trivial to defeat; just write a program that automatically sends a space or two to hold the line. Another way to handle the idle question is to disable the idle timer until nearly all the modem lines are in use. Always remember to treat your users with courtesy. "If you explain things to them the most of time they will go out of their way to make things right." +.# What can be done about users who never log out? What about users who seem to be on the system 24 hours a day, 7 days a week? This leads us to the controversial question of pricing. Historically, Internet service providers have charged a fixed fee per month, regardless of the amount of use made on the system. This has almost always been the best model for customers; all but the lightest users pay less than they would under the non-fixed schemes. Unless they have unique offerings like the slick graphical interface of NYC's Pipeline, new providers are not going to be able to come in at higher fees than (say) Netcom or CRL. This pretty much eliminates the idea of hourly fees for most. Well, maybe. Draper Kauffman (draperk@io.com) notes that Netcom is in his area at their normal rates. Despite having higher rates, his system and other local competitors have not suffered; the reason, of course, is Netcom's infamously terrible service. He suggests that excellent service can still get $ 25/month or more. Hourly fees are mandatory, of course, if you offer continental US toll-free access. This can often be arranged at very competitive hourly cost as compared to a toll call to your site. Some services, particulary bulletin boards, undercut the typical ISP monthly rate but restrict access to a certain amount of time each day. The Pipeline offers a set number of "free" hours and charges for any longer period of time spent online. +.# What about people who stay on their SLIP account forever? Unfortunately for those of us who want to provide unlimited accounts, the growing popularity of SLIP/PPP makes it very difficult to stick to our guns. SLIP/PPP accounts are unobtrusively there; the customer's computer is part of the Internet, and the most natural thing in the world for many users is to just dial in to the system and forget it; use their computer normally and access the Internet when they feel like it. SLIP email programs can be told to check for mail every five minutes; that's way below what any sensible idle timeout would be, so the effect is for the SLIP user to be on the system 24 hours a day, 7 days a week. There are three basic approaches to dealing with this problem: * Have a very long time limit on the account, usually around 150 hours a month. This forces people to keep track of the time they spend on the system. * Have a policy that says that you monitor excessive usage, defined as being online and not doing anything actively other than routine mail checks. People who have used this policy report that most people who are told that they need a dedicated (circa $ 100-175/month) account will get one. * Charge by the minute from the first second of use on. Most users hate this idea. +.# Monitoring Your System If you're a single individual, how can you hope to deal with system problems in an expeditious manner? Bryant Durrell has an answer: "Since you probably won't have someone monitoring 24 hours a day, you'll need some sort of notification of urgent problems short of users calling you in the middle of the night. One solution is a beeper. If you have a spare modem and a beeper, it's possible to write a simple syscheck script that beeps you whenever something goes critically wrong." +.# Trouble Ticket Systems (*) This is an issue for providers that presently have employees, so I will let someone answer this who is in a better financial condition than me. :-) + Who needs and wants Internet Services? How can we reach them? This is definitely an incomplete and imperfect section; additions are especially appreciated here. +.# Graduating College Students Students graduating from college have already gotten addicted to the net and will pay reasonable money to gain it through a commercial provider. Reach them through advertisements in alt.internet.access.wanted and similar newsgroups. Postings on free college bulletin boards are also likely targets. +.# Parents of College Students Jason Goldberg <newleaf@attmail.com> has this fine suggestion. Mom and Dad can keep in better touch with their children while they're away by purchasing an Internet account. "CIS and AOL do a very healthy business among people who just need e-mail access. Consider offering an e-mail only account in order to beat the commercial services on price." Jason suggests that you contact high schools about running ads in graduation programs and yearbooks; emphasize the cheap lifeline this offers between parents and children. "E-mail has proven to be a very effective way for people to stay in touch who would never call or write via US mail." +.# Special Interest Groups The Internet has many wonderful discussion groups on issues ranging from sex to Unix. Jason Goldberg suggests contacting local groups, from chess clubs to dog breeders, and showing them a pre-researched set of resources that can be obtained on the net. For example, chess enthusiasts would enjoy reading rec.games.chess and the ICS (an Internet chess server that lets you battle real-time opponents from around the globe). +.# People with Internet access at work There are many people who would like to use the Internet outside of work. They will pay for an account that gives them privacy from their employers, allowing them to participate in the "forbbidden" sections of the net without fear of sanctions. The internet access wanted newsgroups are probably the best ways to find these people. +.# Current callers to computer bulletin boards (BBSs) Bulletin boards are getting hooked up to the Internet in massive numbers, but the culture clash between them and people with "real" Internet access is very strong. Most bulletin board systems don't offer newsreaders with killfile or threading capabilities, and as a result the quality of access tends to be very low. In addition, direct services like Gopher, FTP, etc, tends to be either non-existant or available at substantial extra cost. Although this situation is bound to change somewhat, my experience is that BBS software vendors are remarkably clueless in this regard. BBS callers are used to paying fairly minimal amounts of money for access, but this seems to be changing; I know of several large systems that charge around $ 15/month, which would get you a shell account from a number of small providers. The best way to compete with BBSs is to ask to be added to major BBS lists. A good guest account or new user routine is a must; BBS callers are used to getting a good taste of what they're getting before they have to pony up the cash. +.# The General Public They are obviously intrigued, if you look at the sales figures of Internet books and information. Certainly there has been a massive flood of Internet stuff, such as the Time cover story, and that has no doubt piqued a significant amount of interest. However, it may be quite difficult to reach them. Flyers posted around bookstores might be a good approach, since people who find Internet books are undeniably interested in access. I'd love a few more ideas to add to this section of the FAQ. :-) +.# Some low-cost marketing ideas (*) The cheapest way to market your services is via the World Wide Web (WWW). With the commercial online services rapidly getting Web page access, more of your potential customers than ever can see your message. And, since people tend to seek out Internet providers in their area, the WWW is an ideal marketing medium. B J Herbison advises that you do the following when designing a web site to promote your system. (I have added some of my own comments to his remarks, so you can consider the following jointly authored). * Find (via Yahoo or B J's own list of ISP listings) all the ISP listing pages you can. See http://www.herbison.com/herbison/iap_meta_list.html . * List the services you are willing to provide in clear, concise language. * Keep your entry up to date. Nothing's duller than a stale Web page. * Review your competition's entries and make sure you measure up. Don't forget any special services you provide that differentiate yourself from your competitors. * Make your own Web page. Don't make it dependant on Netscape; people in need of an Internet provider will call you from AOL, Prodigy, or even text only sites. * Remember that the most important thing about an Internet provider to your customers is whether it's local to you. You might be impressed by MCS in Chicago, IL, but if you're in San Diego, CA, you're not going to be able to use their service! Be sure it's easy for people to know that you're local to them. B J suggests you check out http://www.iii.net/iii-info/local-di.html for a clear presentation of the cold, cruel phone company facts. Draper Kauffman suggests that we contact our local public television station and donate 10-odd subscriptions for auction during pledge week. "They were auctioned off two per night for five nights, got a lot of comment because of the novelty, and went for good prices (i.e. more than we charge). They also generated 30 to 40 paid subscriptions and excellent name recognition in a good demographiccross-section of markets, including some groups we would normally have trouble reaching. "On a guess, if you cound referrals from these new customers, we'll probably get 60 to 100 paid subscribers for the cost of supporting 10 customers for a year. Most important for a startup: there's no actual cash outlay! Compare that to thecost of producing a 60-second TV spot and having it aired 10 times during prime time." He adds that this gambit may be effective exactly once, since his competitors are also going to be doing it next year, and the novelty value will be gone. But for the virtually zero cost, it's a phenomenal return. I might add that public radio might do nearly as well, and many markets have more than one public radio station. So even if this has been done already by another provider, you may want to see if there are stations or outlets they've forgotten. Alicia Salomon recommends the Guerilla Marketing series of books (Guerilla Marketing, Guerilla Marketing Attack, Guerilla Marketing Weapons, etc), which include a number of great marketing tips and tricks. Many of them apply only to storefront businesses, but a healthy percentage of them will work just fine for providers. I have read the books and wholeheartedly endorse her recommendation. Chris Hawkinson added two interesting ideas: > Offer "a low level of free access" to local libraries and schools. "The publicity and good will alone can make it worthwhile. Further, since it isn't full access (let's say just to Gopher), users can quickly get hooked. Lastly, you can use the libraries as an 'information provider' to add more value at the local level. > Offer a reward for new users. Everyone who gets a new user to join Chris' system gets $ 5 for each $ 55 (yearly) BBS account. Jason Goldberg has the following excellent suggestions: * The Internet is one of the hottest things in the world right now, but few know how to use it. Offer and package your expertise by offering one-day Internet seminars to your local colleges, high schools or adult education centres. You'll get a few customers, and you'll also be paid as a teacher as well. Don't miss out on this negative cost marketing opportunity! * Produce an Internet newsletter. Write articles about the latest trends in the net, and the latest hot sites and IRC channels. Include information about your service and a special offer (introductory rates, waived set-up fees, etc) to get people interested. This is, incidentally, also an excellent way to retain customers who might otherwise lose interest in the net; give them something interesting to check out every month, and your renewals should soar! * Produce an Internet column for your local newspaper, free weekly or computer publication. This could have basically the same content as your newsletter. Be sure to include your company name and phone number in the author section. (Example: "David H Dennis is owner and CEO of David's Amazing Internet Services; modem (818) 997-7500"). * Bundle your service, including free setup or a free month, with the purchase of a new computer or modem from local computer stores, or membership in local computer clubs. Present your service at a local computer users' group meeting and raffle off a free year. Here are some general suggestions, also from Jason: Distinguish yourself from the competition; don't make the mistake of focusing solely on price. The more value-added services you can offer, such as an Internet newsletter or seminars, the more likely you are to prosper in this hotly competitive world. Offer monthly seminars, a newsletter, and/or value-added software. Include Internet books, videos or software with membership. Make sure people understand your unique advantages - T1 vs inferior competitors' options, faster computers, more disk space, etc. Focus on differentiating yourself, not on raw price! Set specific goals for your service, and work towards them. "I would like to add 50 new accounts a month." Is it easier (and cheaper) to upgrade existing accounts to SLIP than to beat the bushes for new customers? If you have a lot of former customers, it might be very cost-effective to lure them back with a mailing of a special offer. This is especially true if you can show how you've improved during the time they've been gone. Offer special deals for customers with accounts on other services - CIS, AOL, Prodigy, Netcom, etc. Sell additional Internet products - books, videos, software, other services. Rent or trade your mailing list to local computer stores or other businesses. Consider a co-op ad, where you team up with other local businesses to buy a large ad instead of purchasing multiple small ones. Encourage your customers to call or write mail when they have a problem. The "silent majority" of customers don't complain - they just move to another provider. Be sure you treat the people who do send you mail with genuine concern, and they will award you with loyalty, even if you don't manage to fix their problems. (I [the FAQ maintainer] know this one from personal experience!) Send them a holiday card, thanking them for their business. Give them rewards for referring business your way - i.e. a $ 10 discount on their next subscription per paying subscriber referred. (Jason Goldberg is a sales and marketing consultant specializing in interactive, entertainment and technology businesses. Former Senior Director of Sales and Marketing for Blockbuster Entertainment, and General Manager for an IBM/Blockbuster Technology Joint Venture. He welcomes e-mail with your questions, comments and success stories. He sells a 90-minute video on the Internet for new users; contact him for details and pricing information.). + Internet Marketing A great deal of interest has started to appear about marketing via the Internet and the World Wide Web (WWW). Ever since the first graphical browser hit the streets, the public has been fascinated and the advertising community intrigued. "Internet Malls" have sprung up all over the country offering merchants a low-cost way to display their wares to a massive audience of people all over the world. Members of the Internet Marketing mailing list have debated long and hard over this trend, mostly bemoaning the opportunistic nature of many of these ventures. I've visited a few low-rent Internet malls, and they remind me of a moldering central business district of an insignificant city. A curious stew of products, none particulary distinguished sounding, sold at not particulary good prices. From what I understand, they haven't done much business. Creating a mall similar to them would be good for short-term profits, but don't count on your new 'tenants' staying long. Remember that people are on the Internet for information, preferably interesting, detailed and timely information that changes rapidly. If you want to do an Internet marketing site that people will keep coming back to, here are a few suggestions gleaned from the Inet-Marketing mailing list: * Choose your prospects well. Sell your services to people who really have information to contribute as well as a desire to buck up their sales. The best prospects for Internet marketing are companies whose customers are passionate about their products. A Harley-Davidson motorcycle web site, for example, would work great, because people who use them are very passionate about them and are always eager for more information. A web site for Crest Toothpaste would probably be a flop, because people don't want to devote much energy thinking about toothpaste. Sure, they want good toothpaste, but it's not what they want to spend time reading about. | An excellent example of this is HotHotHot, a Pasadena hot sauce | shop that substantially increased their sales through their Web | Presence. * Change the site often. Make sure people want to revisit it. Having a 'magazine' with weekly or monthly issues is an excellent idea. But remember, the more often your site changes, the more likely people will remember it and come back. So a weekly change is strongly recommended if it's possible. It might work to refresh parts of the sites on a rolling weekly basis, as long as the last change date for each section is made clear. * People are unlikely to visit your site without a strong campaign in other media. The best medium to advertise a Web site in is probably USENET Newsgroups corresponding to its subject. However, it's not generally a good idea to write messages doing nothing but advertising the web site. People are most likely to come to the web site if you are an active participant on the newsgroup, asking questions and sharing your expertise with others. Just put your URL in your signature line, and people will come. Like this: ------------------------------------------------------------------------- David H Dennis * Author, Inet-Access FAQ * http://www.amazing.com Remember that this person isn't necessarily responsible for what he says. ------------------------------------------------------------------------- * Don't rely only on the World Wide Web. It's flashy, sure, but many people have email-only access to the net. You should have an email address for information and comments, as well as a "mailbot" that automatically sends information to anyone who sends a message to a dummy user on your system. * Get on the major WWW indexes, such as Yahoo. Most of them will either have a form you can fill in to add your listing, or a Webmaster you can mail to asking to be added. To increase your chances of being added, be sure to give some details on you, your organization and your relationship to the subject matter. To get on the Inet-Marketing list, send the message "sub inet-marketing" to listproc@einet.net. For information, send the message "info inet-marketing" to the same address. + Now that I have users, how should I deal with them? NOTE: I AM NOT THE WORLD'S GREATEST EXPERT ON THESE ISSUES OR ANYTHING. IF YOU WANT TO STAY OUT OF TROUBLE, BE SURE TO ASK A LAWYER ABOUT ANY DISPUTABLE PROVISIONS HERE. +.# How should I introduce my users to the unique Internet culture? This is really a matter of USENET, electronic mail and mailing lists, since most other resources are purely read-only. I had some long conversations with a number of people heavily involved in net administration, and they asked me to convey the following to new users of the Internet: * Tell them not to post for about two weeks to a month. It takes a while to understand what's going on in the unique Internet culture. Absorb what's going on before beginning to speak. * Read Frequently Asked Questions files (FAQs) before asking for help. There's a wealth of information there. * Read the messages on news.announce.newusers. They give you a lot of great information on how to pose questions properly and how to get the most out of your access. It should also be emphasized that, although the guidelines on mailing lists are very much dependent on the list itself, generally users should conduct themselves therein by following the same basic rules as on USENET. Because mailing lists generally attract many fewer posters than newsgroups, and since the software used to read them is usually more primitive, "flaming" and other anti-social practices can cause considerably more harm to a list than a newsgroup. +.# What sort of disclaimers do I need to protect myself? [legal matter - our legal advisor should answer this] +.# How should I deal with Netiquitte violations? There are as many different opinions on this issue as there are system administrators. The general rule, however, is quite simple: a single complaint can probably be safely ignored as a dispute between users and not true wrongdoing on the part of your user. Multiple complaints should be investigated, and proper warnings given. Only after a user has received complaints about multiple incidents should his account be in true danger. As a general rule, your user is innocent until proven guilty. Many obnoxious groups, such as the infamous alt.syntax.tactical crowd, will complain to you about a perfectly innocuous user in an effort to get her or him kicked off. The law firm of Canter & Seigel was kind enough to tell us all about their Immigration Law services a few months back. Many of us appreciated their help so much that we wrote to thank them personally for the information. :-) Unfortunately, so many people were so effusive in their thanks that they filled up the entire mail disk of C&S' service provider! With the kind of blatant nastiness of C&S, there is really no excuse not to drop them straight away. Indirect did basically all the right things in their response to the C&S incident: * Yanked the C&S account instantly. * Posted messages to news.admin.misc and other appropriate groups telling the world about their response. This helped stop the mail bombs and other attempted punishments of Indirect. * Set up a gopher server with fuller information about the incident. (I welcome additional information on dealing with problem users and such) +.# How should I deal with disputes between users? Unless it looks like they are going to physical blows, you should ignore most of these disputes. Dealing with physical threats is a more serious matter. I have actually had that problem on a BBS I operated, and my course was to ignore them. What I actually should have done is open to debate, and I'd like to hear opinions from others who have faced similar problems. + Wanton Women and Straightlaced Crusaders: Pornography on the Net +.# Introduction: Where pornography and the Net collide. On the television every night, people watch other people shooting at yet other people and those people fall down and ooze blood and often never get up again. A couple making slow gentle love by candlelight is an x-rated movie, condemned by all and sundry, and yet we see more and more people being killed and wounded and hurt on the tv with little moral condemnation. Our world is messed up, and Internet providers are affected, oh yes we are. And it all begins with one disturbing fact: People get on the Internet because they want to take a peek at the dirty pictures, because they want to read about Tammy, the 14 year old horny teenage slut, because they want to express their sexual fantasies and read those of others. So we start with a big gain: All those great Internet services, all the access we make available to museums and research facilities, all that great hobby information and social chatting and Kibology and what-not is subsidized by the legions of people who do nothing more enlightening or inspirational than downloading dirty pictures. The same forces that make the Net useful for getting sexual materials are the same ones that make people fear the net. Normally, you'll get sex stuff by visiting your local porno shop, a place filled with high prices and rather dubious looking people. You wouldn't want your best friend to catch you there, would you? With alt.binaries.pictures.erotica and similar groups, you can see those same pictures in the comfort of your own home, without paying a frightful lot of money, and without much chance of embarassment. In this way, pornography is destigmatized in a way that dramatically increases its consumption. So people flock to their Internet provider because it provides the fun stuff they want, in the most uncensored way. They're having their fun, and we're taking their money, and you'd think it was just another example of good ol' American ingenuity at work, eh? Perhaps not. For one thing, the transmission of obscene material across state lines is illegal in the US, and much of USENET and FTP site material and WWW sites qualifies easily as that. There have been some truly ugly cases. For example, a California BBS operator was sentenced to jail for accepting a subscription for his adult BBS from a Southern state, and thereby allowing the subscriber to download an adult image judged illegal in that state. The situation of alt.binaries.pictures.erotica and other USENET newsgroups dealing with porn is rather different and very unusual. The California BBS operator's system was dedicated exclusively to the distribution of adult images. An Internet provider offers so much material that the intent is much less clear. The provider can use what is called the Common Carrier defence - meaning that he just distributes stuff without looking at it, and so he is not responsible for the material. The common carrier defense is used by the phone company and post office; they say that, since they don't listen in to your phone conversations or read your mail, they cannot be responsible for any material you send via phone or mail, even if they're making money off it. (900 numbers are a primary example of this). This could be logically extended to USENET news; nobody has enough time to monitor every news article written or every e-mail message sent. In the case of the net, though, there have been no court tests of this argument. It might come to pass that we're not responsible , or it might not. In the case of alt.binaries.pictures.erotica, after all, we make a decision to carry it as part of our newsfeed, and we certainly have a reasonable idea of what it contains. I certainly know perfectly well that I am carrying this material on my system, and that it is of an adult nature. I especially know that people sign on to my system for the purpose of reading it! Matthew B Landry was kind enough to dig up at least one precedent that appears to be in our favour. I'm collecting a bit more information on this, and I will issue a revised report soon. Some people have considered all this and put forth the argument that the safest course of action is not to carry the adult material at all. Remove all alt.sex.*; remove all alt.binaries.pictures.*; read through the active file and eliminate anything even vaguely resembling sex. Others note that this violates the classic big company principle of denying responsibility. If you remove alt.sex and its friends from your feed, you are effectively accepting responsibility for what remains; you are saying that you are not a common carrier who just distributes everything received regardless of content. This could be a very dangerous thing to do, because lawsuits will seek the easiest targets, those systems for which this does not apply. So if you remove alt.sex but don't remove news.answers, and if an obscene message (such as the alt.sex FAQ!) appears on the latter, you might get in trouble for it. The worst thing to do is to remove the sex stuff and announce that you have a "child proof" system. Inventive children are rather frightfully good at finding ways around it, especially when USENET is on their side, not yours. (There is a document called "The Reading Banned Newsgroups FAQ", which explicitly lists the ways this can be done). Most people who run USENET sites are very much against censorship of ANY material, whether adult or not. We aren't going to remove alt.binaries.pictures.erotica unless we're forced to do so. It has to be said that this is a fight about money, as well as principle; a large number of our subscribers would quit if they didn't have the binaries and the sex.stories to play around with. For once, principle and making money stand together, on the same side. The Christian Right, unfortunately, are rather unhappy at this state of affairs. Certainly it's a very tough problem from their point of view. They believe sexual material to be abnormal, perverted and just plain wrong. Just as many of us believe in free speech as an absolute, they believe in respect for God and his anti-porn will. And, thanks to the Republican takeover of Congress (which I otherwise support), they have a good chance of winning some kind of tough penalties for the distribution of pornography on the Internet. +.# S.314, Senator Exon's bill There has been a lot of fear and loathing surrounding Senator Exon's new bill, S.314. I was able to get a copy of the actual bill itself, together with reviews of the bill from various sources. A superficial look at the bill's contents might make one wonder what the fuss is all about. The bill's actual content takes existing law on telephone harassment as its base, strikes out the word "telephone" and substitutes the words "telecommunications device". In short, it seems to bring laws on telephone harassment into the telecommunications era. If you are a woman, and I call you and terrorize you with obnoxious sexual expressions, that's against the law. If you are a woman, and I email you with similar expressions, it's not presently against the law. Senator Exon's bill changes that so that use of a computer is basically equivalent to use of a telephone. He also increases the penalties for such violations substantially, from $ 10k to $ 100k. So far, this sounds pretty darn non-controversial. (Well, maybe the new penalties are a shade over-the-top, depending on the exact nature of the communication, but the goal of the bill is surely not unreasonable). The most important question is the definition of "obscene or harassing communications". Does this refer to electronic mail, sent to a single party without her or his consent, or does it refer to someone reading messages on a random newsgroup? Certainly it would be quite easy to think of the contents of some newsgroups as obscene; indeed, they were meant to be that way. Does this group refer to them, or only to communications directed towards a single individual? Certainly the name of the bill, the "Communications Decency Act of 1995", does not inspire confidence on the part of the civil libertarian; it seems to imply that it's cleaning up a lot more than the online equivalent of harassing phone calls. Some people have felt that this bill requires the operator to take a pro-active stance against harassment. This doesn't sound right to me. The full text of the bill does not mention telephone companies or on-line service providers even once; its provisions are aimed at those who initiate the communications. So one of your users might become liable under its provisions, but you as the service provider would not be. Indeed, since the bill equates telephones and "telecommunications devices" by subjecting them to the same laws, it might be a good piece of evidence for the "common carrier" defense mentioned above. I would suspect that the obligations of the service provider would be similar to the obligations of the phone company in a similar matter; you would have to identify the perpetrator by checking your records, or cooperating with another site in doing so. This is something sites already do on a volentary basis, so I doubt that it would mark much of a change. Walter Vose Jeffries points out that S.314 as written does include transmission. "makes, transmits or otherwise makes available any comment, request, suggestion, proposal, image or other communication." He tells me that the bill has since been reworded to exclude common carriers, thus changing some of the most objectionable aspects of the bill. It wasn't clear from his remark, however, whether ISPs and BBSs are considered common carriers and thus outside the bill's reach. However, a vital point is that the bill is completely silent on the question of precisely what an online service provider's duties are in this regard. In theory, they could extend to anything, as little as nothing, and as much as putting together a whole bureaucracy to monitor e-mail. Its job is merely to say that there are substantial penalties to the originators of obscene messages; it says nothing else, including whether it includes public messages as well as private. One of the major questions here is what would happen to the operators of systems running anonymous remailers. They might be required to disclose the identity of people who used their service in good faith to broadcast controversial views. How can the rights of those people be reconciled with those of the victims on the other end? This is particulary interesting in view of the recent Church of Scientology case, where they are trying to prevent anonymous remailers from re-broadcasting CoS material. S.314 is likely to be strongly opposed by the cable industry, since it requires the re-working of set-top boxes to fully scramble the audio (as well as the video) portions of pornographic material sent through cable systems. This is certainly its most clearly understandable provision. If the bill is construed to apply to public as well as private communications, it may face significant hurdles in the courts. Traditionally, the courts have treated issues of nudity, indecency and obscenity separately; this bill attempts to combine them in a way that may fall foul of the First Amendment. The issues of Community Standards, at the heart of US law on pornography, are definitely thorny when applied to the Internet, which has global reach. What is legal in Los Angeles, California is not legal in Fargo, North Dakota; what standard can be used for a worldwide communications medium? At present, there are no guidelines, only guesses. This bill has some valid ideas; surely we can all agree that people who harass others via email should be subject to the same penalties as those who harass by telephone. However, its provisions on electronic communications are murky to say the least. On the whole, it seems like a bill well worth opposing despite its innocuous surface. However, the industry might be able to support a bill that was less vaguely worded. To check out a copy of the bill itself, visit the URL http://thomas.loc.gov/home/c104query.html and ask for bill S 314 by number. This is a very nice search system and is an excellent way to get copies of any legislation that might affect you. For another analysis of the bill, check out gopher://gopher.panix.com/vtw/exon . This analysis echoes most of the my concerns, but has a somewhat more unfriendly view of the proposed legislation. You can also obtain additional information by sending a message to s314-info@cdt.org. +.# Porn and you: The bottom line I think you can tell that this is an amazingly controversial issue, with opinions literally all over the map. Some people think your safest course is to give everyone complete access to everything; others think a good faith effort to prevent minors from accessing porn is needed, and still others think porn is horribly immoral and should be removed from the net entirely. I think most Internet providers believe in their heart of hearts that "information should be free"; most of them would also rather not have their equipment seized by a rapacious Federal government. I suspect the bottom line is something like this: - Someone's going to become a test case on this. Unless you're a big system, worthy of an expensive lawsuit, it probably won't be you. Once the test case has been fought and won (or lost), everyone will no doubt change their system (or not change it) to conform to the result. - The safest and most probable course is probably to do whatever other providers in your area are doing. In my experience, most of them offer a full newsfeed, including the sexual stuff, and require potential users to sign a paper saying they're of legal age or have permission of their parent or guardian to sign on. If you do the same, it's pretty unlikely you'll become a test case. There are juicier targets. - Watch out if you promote your service heavily as having these pictures or stories. People who want them will ask or root around the system until they find them. If you promote your system or make it clear that the binaries groups are a major reason for its existance, I'd anticipate a legal problems, on either the obscenity or the copyright issue (see below for more on that). . As another example, you probably want to make UUDECODE available in your libraries, and make sure people know how to get to the FAQ, but it's not recommended to do the decoding for the customer; that would give you a higher level of responsibility for the images. (As long as you offer UUDECODE and carry image groups other than the sexual ones, you can say with reasonable legitimacy that this is a service to help people decode all pictures, not just pornographic ones). - Stories are safer than pictures; verbal representations of things don't seem to scare off the morality police in the way pictures do. There seems to be a legal presumption that favors printed text as protected speech, even if the text is about things that would be distinctly illegal if done and photographed. So kiddie porn stories are OK (which is good, since alt.sex.stories is filled with them), but kiddie porn photographs are most definitely NOT OK at all. More information on this would be appreciated from the legal folks on the list and elsewhere. +.# Copyright Issues If the issues above sounded too confusing to you, you're not going to like this next section, either. You may have wondered where the images on alt.binaries.pictures.erotica come from. "Surely," you must have thought, "there aren't too many teenage nymphomaniacs begging to show off their hot bodies on the net for free, when they could pose for Penthouse and make big bucks."" Sadly, you are right. Most of the images showcased in the binaries group are scanned from commercial adult publications, or uploaded from adult CD ROMs, many of which are copyrighted. As a result, virtually every post on alt.binaries.pictures.erotica is a copyright violation. Karl Denninger has mounted a holy crusade against these images, mentioning the copyright issue as the basic reason. I don't have any quotes from him handy, but his arguments have the hefty force of a loaded bulldozer, or a COBRA Attack Helicopter. He says that, thanks to various clear court decisions he cites, the alt.binaries.etc images are clearly illegal and you're risking your site and your financial viability as a human being by letting the images appear. And, of course, the common carrier defence can hardly be valid, since these groups are labelled in the "official" group description list as "Gigabytes of copyright violations." It is your moral responsibility, says Karl, to take these groups off your system - as he has himself. Counter-arguments bring up the common carrier defence, as well as a few other interesting facts. (1) By removing alt.binaries.etc as copywrited material, you are risking that people will see other copyright violations on your system and be more inclined to think you took responsibility for them. Copyright violations occur every day in news, with people posting things such as Canter & Seigel articles in news.admin.misc. (2) No copyright holder of sexually oriented materials has yet come forward and attempted to enforce their rights to a USENET poster or site. In one particulary fascinating example, a user of CRL posted a scanned image from a Playboy video to the net. Despite his admission of having done the dirty deed, and his willingness to act as a test case, I don't believe he was ever prosecuted. Megabytes of discussion were created, of course, but hey, that's USENET for you. T Kurt Bond did drop me a line saying that TSR Games has made such an effort. For more information, check out this URL: http://www.amazing.com/internet/tsr-licensing.html . In addition, there has been a concerted effort by some publishers to crack down on people using copyrighted illustrations from the Winnie the Pooh books in their Pooh-related homepages. In all cases, however, it would appear that copyright holders are trying to strike a delicate balance between protecting their works and antagonizing the creators of game or Pooh-related resources; these people, after all, are usually the company's most fanatical customers. For that reason, I suspect adult materials are especially vunerable; if Playboy or some other adult publication cracked down, the loyalty of many of their customers would be substantially eroded. This doesn't seem like a risk Playboy wants to take. Considering that most of the material posted is from back issues whose marginal value is quite low, I can see why Playboy and others would make a deliberate decision not to prosecute. In theory, the First Amendment should support our right to read porn. However, in the copyright issue, I don't think the law is on our side. Practicality may be, however; as long as it is basically the responsibility of individual posters to stay within copyright law, it's unlikely it would be worth Playboy's time to go after them. In the widely cited court cases involving Playboy and BBSs, the BBS had an active role in the creation of the site, in particular doing the actual scans of the images. However, if Playboy could somehow find an Internet provider liable, they'd probably do it. Summary: Although the alt.binaries.etc groups are blatant copyright violations, it's quite unlikely that they'd choose your own site as a test case. As a result, I think it's fairly safe from a legal point of view to carry alt.binaries.pictures etc. The moral issue is rather more interesting. +.# Can minors be allowed to use unrestricted accounts? With all the spicy stuff on TV, in movies and in books, all of which can be easily accessed by minors, it rather surprises me that there's so much fuss about a few pictures. But there is. The answer appears to be, "Only if they have a note from their parents." But I'd like to hear all views on this issue. Note that this is not confined to picture files; newsgroups like alt.sex and alt.sex.stories also contain questionable material. Even such a tame sounding group as news.answers is a time bomb; eventually that alt.sex.bondage FAQ is going to find its way in there. +.# How can I prevent minors from seeing the sex stuff? Software has been created that attempts to prevent minors from seeing sexual material. With one exception - Bess, the Internet Retriever - this software resides on your customer's PC and is supposedly controlled by parents. The main problem with this software is that a clever kid can disable it all too easily; instructions on how to do so for popular blocking programs are freely available on the net. The most important argument against this software is that it attempts to codefy parental responsibilities, which are more a matter of personal taste and judgement than impartial, solid and standard. For example, Solid Oak Software's CyberSitter [ http://www.solidoak.com ] was heavily criticised in many circles for blocking phrases such as "homosexual", "safe sex", and so on. Critics said that this could prevent people who need it most from reading information on safe sex that could potentially save their lives. Solid Oak responds by saying that this information is about immoral topics that should be suppressed. Unfortunately, this is the very nature of the beast; this kind of software is going to be bought by people wanting restrictive controls. But they should be warned that many perfectly innocuous sites can be banned. For example, if they find that Jennifer's home page on my server is not suitable for kids, their recourse is to block the entire amazing.com domain, not just Jennifer's page. Because of that, CyberSitter users wouldn't be able to read this FAQ, which seems like a bit of a pity. I would not advise users to buy this kind of software, because I think the cure is worse than the disease. If you must get into this particular business, I would look at CyberPatrol instead; they seem to have the most reasonable policies towards site blocking. In the case of your USENET news feed, you can certainly restrict a lot of it, but I doubt that you'll get it all. In particular, news.answers contains some sexual material. It would be possible to put together an automated kill-file mechanism that would put any group with the name 'sex' in its title, and hide messages with the name 'sex' in their subject lines. My BBS-like software has an interface that gives another possible answer to this question. With my system, you can effectively lock people in to a specific list of newsgroups selected by the operator. Since there's no shell account, there's no way to fool with the active file or read the groups from the news spool. A normal newsreader has so many security holes that it's not going to prevent the determined 13-year old from getting the access s/he wants. I am very much afraid that protecting minors from sexual material on the net is just as hard as protecting them from sexual material in real life. That is to say, impossible. Even with my BBS-like software, they could still telnet to a site that has those groups. It is a great pity this opens up so many potential legal problems. + Legal Issues THIS SECTION WAS NOT CREATED BY A LAWYER, AND THEREFORE DOES NOT CONTAIN LEGAL ADVICE. YOU SHOULD CHECK WITH YOUR LAWYER BEFORE PROCEEDING. INFORMATION CONTAINED IN THIS SECTION IS BASED ON MY REASONABLE KNOWLEDGE AND BELIEF BUT IS NOT INTENDED TO BE COMPLETE. That said, here are some of the legal issues currently facing providers: COPYRIGHT: Many newsgroups contain blatant violations of copyright law. The most prominent among these are the alt.binaries.pictures.erotica* groups, which often include scans from Playboy magazine and other copyrighted material. Karl Denninger (Karl@MCS.COM) avoids potential liability by not carrying these groups at all, and strongly urges other providers to do the same. Others note that there are frequent equally blatant copyright violations on a large percentage of groups, including the full text of newspaper and magazine articles often typed into news.admin.misc. A CRL user proposed to create a test case of this by bragging about his post of a specific copyright violation to alt.binaries.pictures.erotica, including his PGP signature for verification. As far as I know, no action has been taken against him for this. Certainly, there has been no recorded incident of any action taken against Internet providers or their users for copyright violations against Playboy or anyone else. Playboy did win a case against a bulletin board operator who scanned in images from Playboy and sold them via subscription. SLANDER, LIBEL ET AL: Someone more knowledgeable than I will have to write this section. THE COMMON CARRIER ISSUE: This one likewise. + Books and other Resources Linux Documentation Project (LDP). These are a series of on-line books which can be obtained via FTP and printed. An excellent argument for getting a PostScript printer, since that's the only format that really works well. You can also buy copies of these books and other Linux information as the Linux Bible, which is carried by some bookstores. ORA (O'Reilly & Associates). A publishing company that puts out a number of helpful books on network administration. They are clear and well written, with plenty of clear illustrations and helpful explanations. Unfortunately, since versions of Unix differ a great deal, they may not be completely accurate for your system. They are one excellent argument for using Sun workstations or BSDI Unix, since their examples should conform to thoe systems with almost complete precision. USENET FAQs and newsgroups. This is probably the best source of information on administrating a Unix or Linux site - especially if you're using Linux. The traffic on the Linux newsgroups has to be experienced to be believed, and most Linuxers are eager to help. I'm not sure how good th net support is for Sun or BSDI; I know BSDI's technical support is known to be excellent. + Glossary of Common Terms and Acronyms Note: The idea for a glossary was suggested by Tony Sanders . At present, it's woefully incomplete and additions will be warmly appreciated. +.# Glossary of Internet Service Terms These are words, phrases and acronyms dealing with Internet Services issues. There is a second glossary below for words dealing with Internetworking wholesale issues such as CIDR blocks and what-not. Bandwidth. (1) The maximum amount of information that can be transmitted at any given time. A 56k leased line connection, for example, has 56k of bandwidth. (2) An abstract unit of measure of the load something puts on an Internet connection. "Multimedia through the Internet uses too much bandwidth for a 56k connection." Often used in Usenet to refer to stupid questions or wasteful flames, i.e. "this is all a waste of bandwidth". Client. A program that is run by users on their machine. It issues requests to a server, which is generally located on another system. For example, here's how a Gopher conversation might work: #html Connection Active
Client ProgramServer Program
Send me your menu Here it is (sends menu) |
Receives menu from server |
Displays the menu on the screen
Lets client choose an option
Client chooses option
Client finds out what machine to connect to and asks server on that machine for the data |
Client receives data Server sends data |
Client displays data to user
## As you can see from this rough sketch, this off-loads a great deal of stuff from the server program, so that it can process client requests more efficiently. This also makes the system appear very fast, since requests to page through the document can be handled by the client and run as fast as the client can. This is dramatically different from connecting to the remote system and retrieving the file from the remote system screen by screen; depending on the remote system's response time, this could appear many times slower than the client/server approach. For example, the FTP client takes the user commands to transfer files and passes them to th FTP server. The server then sends the file to the client, which then passes it on to the user as appropriate. Note that in the case of Gopher and WWW, the actual connections are very brief. In the example above, the connection is only active when there is a "|" in the "Connection Active" column. This makes use of the server much more efficient then under protocols such as Telnet, where a connection is established for the length of the session. CSLIP (Compressed SLIP). SLIP with compression for a more efficient connection. See SLIP. ECPA (Electronic Communications Privacy Act). A law passed a few years back that says that all electronic mail cannot be read by the people running the system. Its main concrete achievement seems to have been the placing of a notice on all BBSs saying that there is no private mail function on their systems, despite the continued existance of same on the menu. Flame. An insulting message, normally with little real content. A Flame War is a seemingly endless exchange of such messages. FTP. File Transfer Protocol. This refers to a protocol describing the way files can be transferred over a TCP/IP network, such as the Internet. The program used to implement this protocol is also called FTP. Normally, a FTP program is included with your basic networking software, and little needs to be done to make it work on your system. However, if you want to offer anonymous FTP, which lets users on the Internet obtain files from your system, you are urged to consult a book on such as the O'Reilly TCP book. GOPHER. A relatively simple-minded menu interface to the net, Gopher presents a list of menu options. Each option can refer to a resource in a different location on the Internet. Due to its rather dull look and feel, Gopher has been virtually replaced; you almost never see it anymore. Check out the World Wide Web (WWW) for its replacement. HTML. (HyperText Markup Language). This is the scheme used to design World Wide Web pages. Raw HTML is strikingly similar to the text formatters of old, while lamentably incompatible with any text formatter language I know of. Fortunately, there are numerous tools that can help you write HTML with reasonable efficiency; check out WWW page [[ www page where this stuff is ]] To do well as an Internet provider, you'll probably need to learn how to deal with HTML, since you'll almost certainly want to design your own WWW page to promote your system. HTTP. (HyperText Transfer Protocol). This is the protocol used for information transmitted over the World Wide Web (WWW). InterNIC. The government-funded service, run by a company called Network Solutions, that parcels out IP addresses and domain names. Complaints about slow service have been heard quite loudly in recent months. IRC (Internet Relay Chat). This is a direct interactive way for people to hold conversations using the computer. The level of conversation tends to be pretty low, and rather on the dull side, but many users love it. See the IRC section, earlier in this FAQ. Java. This is a programming language created by Sun. It was originally designed for the development of proprietary information appliances, but with that idea looking less and less plausible by the day, the development team discovered its potential for the WWW. So far, few really compelling Java applications have been created; however, the technology looks truly fascinating. Check out http://java.sun.com for details. LYNX. A non-graphical WWW client suitable for shell or BBS account users. See WWW (World Wide Web). Microsoft Internet Explorer. MSIE is the new kid on the block. To be honest, if Microsoft didn't do it, I would say it's a promising challenge to Netscape. As it is, I really don't want MS to be taking over the Internet like they've taken over other aspects of our lives :-(. Use Netscape. 2.0 blows MSIE away. MOSAIC. A World Wide Web graphical client program, usable only under SLIP connections. See Netscape and WWW. MUD or Multi-User Domain or Multi-user Dungeon. These are 'virtual worlds' created by your fellow users. They resemble the old text adventures, such as Zork, but they are populated by people who telnet into them. People who use them swear by them; people who attempt to casually use them are either hopelessly bewildered or sucked into an all-consuming vortex. See the section on Muds elsewhere in the FAQ. Netscape. The One True Web Browser, this program has a virtual monopoly on the web browser market. The story is well-known: Marc Andressen, stuck at a dead-end $ 8 an hour job coding Mosaic, decided he could do better, got some venture funding, and became the technical God behind Netscape. Marc knows what he's doing; Netscape is the easiest and slickest graphical interface platform I've ever seen. Now, Marc is a wealthy guy; I think he makes something like $ 8 a second nowadays. Some people have said that Netscape's proprietary extensions to HTML are the kiss of death for the language; most people just check out the extensions, mutter an occasional, "Hey, cool!" and use them. After all, anyone can download Netscape, and few people who have switch to any other browser. Netscape's a great product, it works exceptionally well, and what bugs there are rarely matter. An excellent challenge to Microsoft's hegemony over life itself. News, aka NetNews, USENET News, Internet News, etc. This is a messaging system that is one of the most famous and popular parts of the net. See the detailed FAQ section on news for details. PPP. Point-to-Point Protocol. A newer and supposedly better way to connect your site to the Internet via a single serial line. Windows95 has greatly expanded its popularity, since it supports PPP instead of the older SLIP. See SLIP. RFC. Request for Comment. This is an informal system for proposing Internet standards. The technical people who work on the Internet upload RFCs to the NIC, where they are given a number and published. Many of them are later adopted as Internet standards. Search Engine. As the World Wide Web has grown bigger and bigger, programs have been created that wander the web, looking for resources of interest. They then put them in enormous keyword dictionaries and let you search for what you'd like to find. The best search engines are Digital's Altavista [ http://altavista.digital.com ], Lycox [ http://www.lycos.com/ ] and Webcrawler [ http://www.webcrawler.com/ ]. In practice, Webcrawler is best if you're searching for something fairly general; if you're looking for an obscure or difficult topic, Lycos and Altavista are both excellent. See also Yahoo, which is really a directory, not a search engine. Server. A program running on a remote system that provides information to a client. See Client for a detailed explanation. SLIP. Serial In-Line Protocol. This is one of several ways to attach a computer to the Internet via a simple (and cheap) modem connection. See the earlier discussion on connecting your system to the Internet for additional information. TCP/IP. Transmission Control Protocol/Internet Protocol. The protocol used to send information through the Internet TELNET. Telnet is a program that lets you remotely log in to any other system on the Internet (assuming you have access). In the olden days, university systems used to give guest accounts on their mainfame computers so members of the public who could afford a modem and computr could take a peek. Unfortunately, due to the higher level of obnoxiousness of current Internet denizens, this has become far less common. People can, however, use Telnet to connect to various bulletin boards, MUDs and special interest systems hooked up to the Internet. Trumpet Winsock. A truly ghastly program, it's also one of the few ways to get graphical SLIP/PPP connections to work under Microsoft Windows. The user interface, however, leaves something to be desired. Something? I should say, "Everything." Ugh. But it does apparently work better than its competitors, if you can ever get it set up right. Windows95. If you've been living in a cave for the last year, you might not know that Windows95 is Microsoft's long-awaited successor to the painfully horrid Windows 3.1. Windows95 is not the be all and end-all of operating systems; Macintosh users should stick with their Macs, and Unix users are going to laugh out loud behind their 21" monitors. However, it's a lot better than Windows 3.1, and setting up an Internet connection is far easier than under Trumpet Winsock (which see). If you need to give people assistance in configuring a Windows95 system, the advice from http://www.windows95.com/ is invaluable. WWW (World Wide Web). This is probably the best Internet browsing system - certainly the most fun one to use. It's not as simple to use as Gopher, but it provides many more interesting options, including integrated graphics if you use the Mosaic browser. See the detailed section on the Web earlier in this document. Yahoo is a comprehensive directory of WWW resources. It's different from the search engines in that you submit a listing to them and a human being checks it out and categorizes it for you. Because of this, a search thorugh Yahoo will almost always find resources relevent to your topic of interest; however, it's less all-encompassing than the automated search engines (which see). +.# Glossary of Internetworking (Wholesaler) Terms You will see these terms a lot, but for the most part you won't need to deal with the actual concepts behind them. IMPORTANT NOTE ABOUT THIS SECTION: This is one of the most horribly confusing things about being an Internet provider. Unless you're wildly successful, you'll never have to deal with any of this stuff. This section is not necessarily accurate; I recieved some contradictory answers to my questions and had to rely on gut instinct to figure out what was what. Corrections and amplifications are warmly encouraged. ASN. Autonomous system number, used in BGP4 routing. If you're just starting out with a single connection through a single provider, your provider will take care of all your routing for you. Once you decide to get multiple redundant connections (i.e. a T1 through Net-99 and another T1 through ISI Network Associates), your system will have to decide which provider you want to route through. This is done via the BGP4 protocol. What this means in practice is that you need to get a lot of memory on your router, and you'll really have to understand how routing works. The ASN itself is a unique identifier for your network, which can be used to direct requests thereto. Once the requests are in your network, your own network software takes over and delivers the request to the proper machine. ATM (Asynchronous Transfer Mode) is one of the most confusing concepts presently on the net. I've sifted through about a dozen conflicting explanations in an effort to get the truth to you, our loyal readers. So first, relax: If you don't understand it, you're not alone. Many actual experts don't understand it. ATM is a datalink-level protocol. This means that it's a way that multiple switches can communicate with each other, regardless of the actual data being transmitted. Think of a regular IP packet as a letter within an envelope. ATM is a process where that letter gets put in a bigger envelope, more stuff gets scribbled on the side of it, and it goes its merry way to the end. Once it's at the end, the original IP packet is removed from the envelope and sent to the IP host on the other end. So the hosts think everything's being done in IP, even though it isn't. Why would one do such a silly-sounding thing? Because IP itself is not well suited to applications like full-motion video, which need a steady stream of small packets and reserved bandwidth in which to run them steadily. ATM will work very well with full motion video, and it encapsulates IP (a fancy term for putting it in the bigger envelope), so it can push around IP packets without knowing what they really are. According to Ehud Gavron , "It's greatest strength is supposed to be its use of fixed-size cells (48 bytes). Fixed sized cells are easier to fast-switch than variable length packets due to inherent efficiencies in DMA buffer transfers. However, the less than ideal size was achieved by mixing the European recommendation for voice usage (32 bytes) with the US recommendation for digital data (64 bytes). As it is, 48 bytes carries a non-compressed IP header, TCP header and up to 8 bytes of data. Extra processing work is done to squeeze more data in." As you might guess, ATM is really suited only to extremely high speed links - 45MBPS or greater. In fact, many ATM systems are planned for connections more than three times this speed! CIDR () Block. In the old days, where men were men, women were women, and the Internet was just a little baby, you would go to an organization called the InterNIC to get your IP addresses. The InterNIC would look in its collection, dig up an address, and send it back to you. There were two problems with this approach as the Internet got bigger. First, each address did not carry any indication of its location with it, which meant that each Internet provider and major gateway had to have an entry in its routers for it. Second, as the Internet got bigger and bigger, issuance of addresses from the InterNIC got slower and slower. Thus, the CIDR Block, a group of consecutive IP Numbers handed over to a single provider. Let's say you used Net-99 as your provider. Net-99 has a CIDR block obtained from the InterNIC. They issue you an IP number from the block immediately, so there's no delay to get your number from the InterNIC. When someone wants to get your address, it looks up it in its master list, containing CIDR blocks, and finds that you're a part of Net-99's. It then knows to speed the packets to Net-99 without any further processing. Net-99 then handles delivery of the packets to you. If we didn't have CIDR blocks, the first site would have to look up your full IP number in the list, which would require a large list containing every IP number on the planet. Because Net-99 does the final resolution of the address, that would be needless lookups and thus slower network operation. What would happen if I changed my provider from Net-99 to ISI Network Associates? ISI has another CIDR block, so the easiest step from their point of view is to give you a different IP number within their CIDR block. This might create considerable confusion as you were switching providers, because some people would know you as a Net-99 client and route you to the wrong place. Eventually, the information would filter through all the net's name servers, and all would be well. What if you wanted to keep your original IP number? There are ways to bypass the CIDR block system in that particular case, although the providers on both sides would rather you not use them. In all these cases, your site itself doesn't need to know anything about routing, other than how to get from your system to your provider. Let's say you grew to a large size and could afford multiple T1 connections to the Internet. Say you got an ISI connection and a Net-99 connection, thus giving you two redundant paths to the Internet. This is called a "Multi-Homed" connection, and requires that you get an IP number from the InterNIC, since you are not in the CIDR block of any single provider. [I'm not positive of that last statement; elaborations would be helpful]. A multi-homed connection requires that you run BGP4 (see its separate entry), which lets you manage all routing yourself. Running BGP4 on a multi-homed connection requires a router on the level of a CISCO 4000-series or better, so it's not for the faint of heart or light of wallet. But then again, neither is getting multiple T1 connections. IETF (Internet Engineering Task Force). They are responsible for Internet standard creating and engineering/design issues. For example, they are working on IPng (Internet Protocol/Next Generation). Not part of the Internet Society. ISOC (The Internet Society). According to Ehud Gavron , "The Internet Society is a private for-profit organization which has nothing to do with the running of the Internet. You might consider it as one big Internet glee club or fan club. Its actions are purely advisory and it is merely coincidental that any of its members are also members of the IETF and other organizations that really do promulgate Internet standards." Their WWW server lists a lot of papers and symposia they're responsible for. See http://www.isoc.org for more information. MAE-East. An interchange point for network providers where they can peer together, run by Metropolitan Fiber Systems (MFS) in Washington, DC. Highly successful; pretty much everyone seems to be on it. It's located on the East Coast, creating the odd West Coast phenomenon of packets that start in Los Angeles, head off across the country to MAE-East and the cross the country again to land in the Bay Area. Some people consider this revenge on the CIX router in the West coast, where traffic from the East coast would whiz over to the CIX router, go through it, and then go back East to its destination. Absurdities like these are amazingly common in routing. MAE-East+ is coming; hopefully it will ease congestion problems currently affecting the present MAE-East. MAE-West+ is also on its way. If the NAPs ever amount to anything, their ultra-high speed connections may eventually eclipse MAE-East. Multi-Homed Connection. This means you use more than one wholesaler for your Internet connections. The good news is that you get greater reliability this way, since you can route through your second provider if the first fails. The bad news is that you open up a big can of worms in dealing with things like CIDR blocks and BGP4 routing. For more information on this, see the CIDR Block section, above. Multiple Interconnection Points. National providers such as Sprint, MCI and Net-99 are encouraged by the NSF to have multiple interconnection points across the country and around the world. This lets them provide more reliable service; if one of the points goes down, packets are re-routed through the other available paths. This way, the Internet can continue to operate, even if there was (say) a nuclear attack on Washington, DC. NAP. Interchange points for network providers set up by the government to replace the NSF backbone. According to Steven N Goldstein , "NSF set them up to ensure that the network service providers (NSPs) that provide national transit for the NSF regionals would be multiply interconnected. That much was required or NSF [decreasing] subsidy to the Regionals ... In addition, the NAPs were intended to be open to other service providers, *at their option*, or interconnection. There are plenty of other interconnect alternatives for those who are adverse to NAPs." Well, that was the intent, anyway. According to Ehud Gavron , "The NAPs are better than MAE-East because they interconnect using the vBNS (very high speed backbone network service) at OC-3 rates (156GBPS over SONET). The NAPs are worse than MAE-East because they aren't up, the interconnections aren't up, there is no vBNS and it's not running yet." NACR. Network Address Change Request. A Network Add or Change request form, used by ISPs to request MERIT to update its Policy Routing Database (PRDB). NCARs have to be acknowledged by the ASs listd as routing from the backbone to the home AS hosting the network. SWAB. Another interconnect, this one between AlterNet and PSI. + Resources +.# Internet Mailing Lists inet-access@earth.com (Email inet-access-request@earth.com) This mailing list is a way to exchange tips, information and questions with other ISPs. This FAQ was compiled with assistance from inet-access members. small-providers@netheaven.com (Email small-providers-request@netheaven.com) This list is a low-volume list that's meant to fulfull basically the same function as inet-access. linuxisp@lightning.com (Email linuxisp-request@lightning.com) This list is for people using Linux as the basis for their Internet provider. freebsd-isp@irbs.com (majordomo@irbs.com) This list is for people using FreeBSD as the basis for their Internet provider. com-priv@psi.com (Email com-priv-request@psi.com) The name's supposed to stand for Commercialization and Privatization of the Internet, but it's really the battle of the titans. Watch in awe as the CIX self-destructs and hear paranoid (but often accurate!) speculations about what the telephone companies will do next. "Just because you're paranoid doesn't really mean they're not out to get you." Monitor the paranoid world by subscribing to com-priv. Or don't, and get a good night's sleep. WARNING: This is an extremely high volume list. I have been told it's virtually impossible to get off of once you're added. [Walter Vose Jeffries was kind enough to provide a correction of my quote and an attribution. "Quote from Henry Kissinger: 'Just because I'm paranoid doesn't mean everyone's not out to get me.' (I think that is the original source of this in case you want to attribute it)."] +.# Internet Newsgroups alt.bbs.internet Newsgroup This newsgroup occasionally contains discussion of similar issues to com-priv and inet-access, but it's mostly people advertising their Internet sites or looking for same. This is a good way to advertise and/or scope out the competition. alt.internet.access.wanted Newsgroup Pleas for Internet access. Once your provider is set up, you should monitor this group to find people who might be interested in its services. alt.internet.services Newsgroup Still more pleas for access and advertisements of access providers. alt.culture.internet Newsgroup Assorted discussions of Internet related issues. Alas, not as interesting as it sounds. alt.internet.media-coverage Newsgroup Interesting discussions of media coverage of the Internet. Low- traffic. alt.internet.talk-radio Newsgroup Internet talk radio and sound bytes. +.# USENET Oriented News Groups news.admin.* alt.current-events.net-abuse These newsgroups contain information about the current flood of "spamming", e-mail abuse and other topics relating to net administration. news.answers This is a repository for all FAQ (Frequently Asked Questions) lists. A fascinating place to browse; be sure to make it easy for your users to access it. news.announce.newusers This contains information all new users should read, especially if you haven't produced your own versions of it personalized for your system. +.# FTP and World Wide Web Sites Inet-Access Mailing List FTP Site FTP to: earth.com:pub/archive/inet-access This is the official home of the latest Inet-Access FAQ - this document. It also contains archives of the inet-access mailing list. WWW Service Providers List URL: ftp://ftp.einet.net/pub/INET-MARKETING/www-svc-providers This document contains a list of known WWW service providers, and is an excellent way to scope out the competition. It was compiled by Mary Morris +.# Books The O'Reilly and Associates "Zoo" - the books with the funny animal engravings on the covers - are universally popular, and with good reason. Some of the information, however, is specific to Unix environments the authors are familiar with. "System Performance Tuning", for example, goes into detail about BSD and System V performance measuring utilities, but little seems applicable towards Linux. Expecially recommended among the "Zoo" is the book 'Sendmail' by Brian Costales et al. Although the Unix-Haters Handbook gulps at its size and its incredible level of detail (it explains concepts anyone running a Unix system should already be familiar with), it's definitely a first-rate introduction to an otherwise confusing beast. The Unix-Haters' Handbook, by Simson Garfinkel, Daniel Weise and Steven Strassman, is excellent comic relief for any Unix system administrator. It tells us very clearly why we should have all bought LISP machines. Or VMS. Or ITS. Or TOPS-20. Only one caveat: Before taking their idea (that "Unix must die") seriously, realize that a similar listing of botched design decisions could have been written for any operating system currently in existance. Don't start me on DOS, Windows or Windows NT! For Linuxers, the books produced by the Linux Documentation Project are first-rate. Fortunately, O'Reilly is apparently going to publish some of them, hopefully giving them a wider distribution. You should definitely look for the Linux Network Administration Guide, which was far more helpful than any other book I've seen in teaching me how to set up my two-station network (Sun 3/60 to Linux PC) and, later, my SLIP connection. If you're cheap and have a fast PostScript printer, they are also available for anonymous FTP from sunsite.unc.edu. Check out the directory (if I recall correctly) pub/Linux/docs/ldp. I don't recommend trying to read the documents in TROFF or TeX format, a true exercise for the machochist! + Conclusion We're in an exciting world here, the emerging world of global communications some call "Cyberspace". Those of us who work to shape this world in our own image are called Internet Service Providers. It's an awesome responsibility. Let's try to make it work. I hope you have found the information in this FAQ to be helpful and of interest. However, to make this document truly useful, I need your contributions, which will be credited as requested. David Dennis david@amazing.com APPENDIX A: Simulating a Router Kevin Smith was kind enough to forward a message he saw on USENET about using BSDI as a router. I hope this will help those who have this problem in the future. This information is reprinted through kind permission of the author, Michael Galassi . I dropped him a follow-up line asking permission to reprint, and asking for a price for the RISCom/N1. He tells me it's "under $ 500 but I don't remember how much." You can contact SDL Communications at (508) 238-4490 for more information and current pricing. Tony Sanders offers the following as an update: "These days you will probably be getting a RISCom/N2 card from SDL, they come in single- and dual-port versions (very nice for setting up a hub). You'll need a driver from BSDI for it as it has a different interface from the N1 card (just ask support@bsdi.com for the N2 driver). Everything else is pretty much the same." This is the last item in the FAQ due to its incredible length (over 400 lines). If you are not interested in this specific subject, you can stop reading now. From: nerd@percy.rain.com (Michael Galassi) Newsgroups: info.bsdi.users Subject: RISCom/N1 summary of experience Date: 18 Jul 94 22:13:34 GMT Organization: University of Illinois at Urbana Lines: 404 Approved: Usenet@ux1.cso.uiuc.edu Message-ID: Reply-To: nerd@percy.rain.com NNTP-Posting-Host: ux1.cso.uiuc.edu Originator: daemon@ux1.cso.uiuc.edu Hi, A week or so ago I requested tips & experience of the list to help me in an impending conversion from a MorningStar Express to a BSDI box containing a RISCom/N1. I've received several requests for this info and two people have provided me with some help. Additionaly, when I ran into hardware problems the people at BSDI were as usual quick to respond, very competent, and nice to deal with (they didn't even sneer at me for not recognizing my hardware problem for what it was). Great bunch. The end result is quite nice, performance is great, the interface to the router is what I'm used to in my normal syadmin chores rather than some vendor's idea of what is best, and the machine is truly flexible, I want more ports I go to any PC store, buy the board, build the kernel and I'm off and running. For starters, the responses I received pointed me at a document available on world.std.com (now ftp.std.com?), I failed to locate the doc there but both people were kind enough to provide me with a copy of this document. I've attached a copy of it at the end of this message. First off, the pitfalls I ran into. o My motherboard (noname vlb 486/33) which appeared to work nicely in an enviroment including 3 ethernets and a DigiBoard routing SLIP did not work well with the RISCom/N1, the symptoms were that within no more than an hour of starting the rn0 interface would lock up, all else seemed OK. o The weather out there was nice and I would have rather have been on my motorcycle. o getting the IRQs, I/O addresses, and memory mappings right for all the boards was hellish. The solutions involved replacing the motherboard, buying a new bike the next day as a consolation prize, and taking very carefull notes *ON A MACHINE THAT DOES NOT RELY ON THE ROUTER*. The basic steps involved: Create a config file rn0 defined. CAREFULLY pick IRQs, memory & I/O locations that don't clash and update your config file. At this point you need to know if your IP provider supports CISCO's HDLC or PPP, the later is more likely. Add a line that reads "options CISCO_HDLC" or "options PPP" as needed. Build your new kernel and install it. At this stage DON"T change any files in /etc such as netstart, rc.local, hosts, or DNS configuration, you'll get that later. Configure the switches on your N1 to match what you've set in the kernel Shut down your machine and put the N1 in it. Boot up ms-dog and run the diagnostics for the N1, the program is called n1.exe and is in \dos on the floppy you got with your N1. There is aparently more than one version of this program for mine did not match the documentation that came with the N1. The program is close enough to "self documenting" that you should be able to wing it. One pot-hole I fell into, I configured my N1 for IRQ 10, same for my kernel, but, the diagnostics won't let you use that interupt. Since I wanted my board to run with the same config as the diagnostics had tested I changed my config a bit. I would guess the longer you run the diagnostics the better, I lost patience at about 30 minutes . When you are bored with watching the diagnostics, power down and attach the cable from the db25 on your N1 to whatever DSU you have, this should cause some LED activity, diferent DSUs will work diferently. Punch reset and watch the boot message carefully, you should see a line looking somewhat like: rn0 at isa0 iobase 0x220 irq 3 maddr 0xe0000-0xeffff Obviously your numbers may be diferent from mine. If you don't see this run strings -a /bsd | grep "rn%d", if you get no output you messed up building the kernel. If you do get output cd to your build directory and look at ioconf.c, make shure the numbers on the line that ends in /* rn0 */ match what you gave in your configuration, if they don't rerun config, redo your make depend & your make, reboot and try again. If they do match they you have a hardware conflict or an incorrectly configured board. Fix and repeat as needed. Once you are up in multiuser mode log in as root, and manualy ifconfig your board with the local and remote addresses, netmask and link flag, I use: ifconfig rn0 inet 199.2.108.234 199.2.108.233 netmask 255.255.255.240 link0 that is, the local side is at 234, the remote end is at 233, the netmask is 0xfffffff0. The link0 means I run PPP on this link, its absence would indicate CISCO's variant of HDLC framing. You should now be able to ping the remote end of your link, in my case "ping 199.2.108.233", if this is the case, you can add a line identical to the one you typed above to your /etc/netstart and you are done! This leavs a *small* matter of routing. If you are a leaf node you can get away with "route add default ", otherwise you will need to configure gated to do the "right thing", this is left as an excercise for the reader. Hope this has helped some, writing it down certainly helped me see what I had done and realize what I'll do diferently next time. If you run into problems doing any of this, drop me a line, I'll be happy to review config files and other similar things to help you out. -michael ---cut here--- > Using BSDI as an Internet Router This document describes the basic procedure for using the SDL N1 board in a BSDI machine to implement Internet routing functionality. This includes useful general information such as N1 setup which should be applicable in many proprietary (non-Internet) network connections. > Introduction The good news is that getting the N1 to work is almost as simple as plug'n play. Additionally, my own tests and from talking to BSDI folks, confirm that even a 20mhz 386 BSDI machine has more enough cpu power to move lotta packets; I get better consistent thruput using BSDI/N1 than I did using a NAT router (and no longer have to deal with a number of connection killing bugs that plague the NAT router). Note that typically when you dedicate a machine to something as important as routing (and other site services such as mail forwarding, POP account, DNS, proxy ftp) don't give people login shell accounts on that machine. Also, for security reasons it might be a good idea to chmod 400 on the /dev/bpf* devices (or disable access to those devices completely once you've debugged the setup). > The Environment The Internet connection is through a 56kbs leased line (PacBell ADN - California) terminated using a Dowty DCP3080 CSU/DSU. The N1 board connects to the CSU/DSU throught a V.35 interface using the cable supplied by SDL. The host with the N1 board then gateways to other machines connected via ethernet. Installing The N1 An overview of the steps involved: o Determine available base I/O, interrupt, dual port ram of your bsdi machine o Figure out if you have to do anything special about caching o Test the N1 to verify setup o Build a new kernel o Boot new kernel; basic N1 test > Setting base I/O, etc. The default BSDI N1 setup requires that the board's base I/O be set to 0x220, uses interrupt 5, and assumes dual-port usage at 0xe0000; these are very reasonable defaults but check your config to see if they will work for you. Note that since the N1 is a 16-bit device, it requires a 128kb segment of dual-port ram (see N1 manual). To check for suitability of preceding defaults, execute the 'dmesg' command to display your machines config. The dmesg command will list all devices base I/O (iobase), interrupts (irq), and dual-port RAM use (maddr). If there are no conflicts, set the boards S1 switch to use base I/O 0x0220 and interrupt 5. The maddr range is set in the BSDI kernel config (as described shortly). Install the board in selected machine. > Setting up for Caching Now determine if the machine with the N1 uses an external cache; if it does, determince if it's a write-thru or write-back. Check your motherboard manual. Higher quality and most newer motherboards use the write-back cache; with this type of cache you do NOT have to configure the dual-port RAM area as non-cachable. Older and cheaper mother boards use the write-thru cache; with this type of cache you MUST configure the N1's dual port range as non-cachable. Depending on other boards you have installed (ethernet), it is possible to run out of non-cachable regions in which case you'll have to turn off the caching. Some motherboards support both type of cache; determine which is enabled on your board and act as needed. > Testing the N1 Boot the machine with DOS and run the N1.EXE test program (provided by SDL). This program will verify basic functionality such as base I/O settings and interrupt. The most important thing this program will do is verify that the desirable dual port ram range works on your machine. The default 0xe0000 address should work fine with most newer MBs using write-back cache; however, older boards might have problems. For example, the machine I installed my N1 in would only pass the dma test at address range 0xa0000. Address range 0xa0000 is also used by vga driver; since I don't run X/have a vga card on this machine, when I rebuilt the kernel, I deleted the vga driver. If you use an older board, be aware of special situations like this. Bottom line is, the N1 must pass the dma test at the memory range you plan on using. If it doesn't pass - don't go any further; things will not work. > Building the Kernel In configuring the kernel for the N1 you should disable any devices you don't need since adding the rn0 device could result in a kernel that breaks certain mem size limits. When you enable the rn0 device, be sure to also change (if needed) its port and iomem parameters. In addition to enabling the rn0 device, be sure you also do the following: o Enable the network option GATEWAY (this enables IP forwarding as needed by an Internet gateway machine). o Talk to your Internet service provider and find out what type of protocol their routers expect. Almost 100% certain that they will say PPP. If so, enable the PPP option; in this case you do NOT need the CISCO_HDLC option (use this option only if you need it). Follow the BSDI instructions and build a new kernel with the rn0 device enabled. > Boot New Kernel; Basic Testing Prior to booting with the new kernel, you should connect the N1 to the termination equipment you plan on using, and turn on that termination equipment. Reboot the machine; you should see rn0 in the device list. If you miss it, use dmesg to verify that the kernel found the N1 at the desired base I/O, interrupt, and maddr range. Once the machine reboots, the CSU/DSU RX data light should be on. Now use ifconfig to enable the rn0 interface. Note that if the N1 isn't connected to line termination device (CSU/DSU), or if cable is bad/incorrect, you'll get "rn0 timeout" messages displayed on the console. Use the following basic ifconfig line to enable the interface: ifconfig rn0 inet Host_IP_Address RemotePort_IP_address At this point the TX data light on the CSU/DSU should come on. If it does, basic functionality is OK, but don't try much else until you read the following sections. At this point you can use two IP addrs from your assigned block, however, the CSU/DSU should NOT be connected to the leased line. I used a Dowty DCP3080 CSU/DSU; the only setting I had to change was to enable the V.35 interface (instead of the serial interface). Note that in making this type of change, you'll probably have to power cycle/reset the CSU/DSU. Note that the rn0 driver doesn't seem to support DTR, so the CSU/DSU DTR light doesn't come on (and neither the DSR light if CSU/DSU configed so that DSR follows DTR). This is not a problem; things will work just fine (at least with my equipment). Setting Up BSDI as Router ========================= The basics steps are as follows: o configure the rn0 interface using ifconfig o test connection to rn0 interface o probably want to get gated as IP router daemon o set default route on other hosts > Ifconfig/Basic Test To configure the rn0 interface using ifconfig, you'll need the following Internet connection parameters: + the IP address for the rn0 interface + the IP address of the interface at the service provider's end + the netmask and broadcast values for the rn0 interface The IP addresses, netmask, and broadcast values you get from the Internet service provider, or if you have an existing Internet connection, you can login to your site's router (if you have the passwords) and dump the config data. If you have no idea what this means, get the info from your Internet provider. The above parameter's are crucial to proper connection function. Don't try anything until you have them; things will not work. For example, my ifconfig (in /etc/netstart) looks as follows (NOTE link0!): ifconfig rn0 inet 131.119.67.134 131.119.67.133 link0 \ netmask 255.255.255.252 The first IP address is the IP address of the rn0 interface (mentioned as $hostname in template form in /etc/netstart); the second IP address is the remote port (of a Cisco router in this case); in /etc/netstart rn0 template this is the __remotehost__. Though it might be nice to use names instead of IP addrs, you'll probably have trouble with names (known bsdi isssue). If the connection is PPP, you MUST specify the "link0" interface option! If the connection is CISCO_HDLC, you don't need link0. Netmask is per service provider instructions; broadcast uses default 131.119.255.255 which is fine, again per service provider info (see ifconfig man page). Before editing /etc/netstart, enter the ifconfig command manually. Verify TX data light goes on in the CSU/DSU. Don't be to anxious; if routed or other router daemon is running, kill them before entering the ifconfig command and verify that your routing table is minimal (netstat -r or -nr). Now create a default route (route add default IP_addr); use the IP address of the machine with the N1. Once the default route is created, you should have connectivity. Test DNS resolution, etc. Things should work fine. If not, use tcpdump to view activity (tcpdump -i rn0). Note that from this point on exactly what happens depends a lot on you Internet service provider. At a minimum I would hope that tcpdump would show RIP requests, and maybe SNMP requests. This would indicate that your connection is functioning and accepting packets. Outbound packets can be verified with something as simple as a ping; this will show that IP forwarding is working. See next section for some Internet related details. If things worked, you can edit your /etc/netstart file and add the ifconfig line. However, at this point, you should comment out any 'route add default' command; see next section. > Router Software To turn the BSDI/N1 machine into a router, you probably need router software. Check with your service provider. They might be willing to set static routes for your site (though don't count on it). If they require true router capability, find out what protocal they want to use. BSDI comes equiped with routed. Routed supports RIP; a very common PPP link router protocol. However, I was not able to get routed to work for my site, and at the suggestion of my service provider, I was told to use gated. There's good and bad news about gated. The bad news is that it don't come with BSDI 1.0; the good news is that it's easy to get and compiles totally clean and simple. Also basic config is trivial, and gated supports a wealth of router protocols. I initially tried routed. The problem was that routed would not respond to RIP requests from my service providers router. Hence, packets from my site would go out, but responses never came back. Specifically, packets originating on the gateway machine (the machine with the N1) would have the correct source IP address and everything worked fine from that machine. The problem was with the other hosts using the gateway machine as the default route. Other hosts packets went out; replies never made it back. Fun to verify this using traceroute and tcpdump. My service provider would not help with routed; they are familar with gated and basically said "use it". Here's my most favorable experience with gated: 1. ftp to gated.cornell.edu; cd pub/gated 2. get gated-R3_0_2.tar.Z (make sure set transfer mode to bin!) 3. Uncompress/untar; read README; follow instructions to build 4. Use the minimal gated.conf file that says "rip yes ;" 5. Install gated.conf in /etc; gated binary in /sbin (NOTE: as built gated is HUGE; might not fit in /sbin on root partition. Do a 'strip gated' to remove symbolic info and reduce to reasonable size.) 6. Edit /etc/netstart to say NO to routed 7. Edit /etc/rc to enable gated Once I rebooted with gated; everything worked! NOTE: Gated README file cautions that for RIP to work, kernel must support UDP checksums. By default, the BSDI 1.0 kernel does support UDP checksums; its all set to work. To finish up, set the default route on all other hosts to point to the gateway machine. Note that if you're switching from a router to a bsdi machine, you could use the IP addr of the router's ethernet interface as the IP address of the bsdi machine. I did not this this so I coauld bounce back and forth between router connection and bsdi connection until things were debugged. > Summary Hopefully the basic steps in this document are useful, however they are not a susbstitute for common sense. Be creative, especially in initial stage. For example, as long as CSU/DSU not connected to Internet line, and ifconfig with junk IP addrs could show that things are basically functional by causing the TX data light to go on. Also keep in mind a cooperative Internet service provider is needed. Good luck in this sense. -michael -- Michael Galassi nerd@percy.rain.com