Archivist's note: Original location is http://web.mit.edu/pmitros/www/lsa/. I'm maintaining this copy because it's an important story in the history of GNU/Linux, which ought not to get lost accidentally.

Linux Standards Association Clearinghouse

This site is no longer maintained. The LSA is effectively dead. A number of other events have happened since this site was last updated that I have not had the resources to keep track of:

This site now only exists for historical purposes.

This site is meant as a clearinghouse of information about Mike McLagan, the Linux Standards Association, and the recent debates. If you have any information, e-mail conversations with Mike, editorials, or anything else you would like me to put up, please e-mail me at pmitros@mit.edu. Please include everything as plain text in the message. I don't have time to respond to all e-mails, but I do read all of them. I reserve the right to publish anything sent to this address.

Update: Internic apparently suspended www.linuxstandards.org, the main LSA site. It can still be accessed at http://199.184.252.30/. I don't know any details about what happened yet. (Oct 10, 98)

The story

On August 18, 1998, a press release appeared on slashdot.org about a new Linux standards organization called the LSA. Although the press release was sent by Lucy Kendall, the organization was quickly traced back to Mike McLagan, maintainer of www.linux.org.

A debate followed. Most people were quite critical of the new organization, for a number of reasons (some valid, some invalid):

Much of the debate quickly deteriorated into emotional and irrational accusations and speculation. However, the most common view seemed to be that LSA had to be stopped. A secondary view was that GNU/Linux was an open community, and anyone could do whatever they wanted (including starting a closed standards group). A common response was that they still didn't have right to the 'Standard Linux' trademark, anymore than the fellow in Boston had the right to grab the 'Linux' trademark a few months back. One or two people viewed the LSA as a good thing.

In response, Mike has posted an editorial on slashdot, which met with more flameage. Ian Nandhra, of NC-Labs (one of LSA's charter members) posted another one on freshmeat, which addressed the need for standards, but did not address the problems with the LSA. The comments section from the referring slashdot article had little flameage, beyond stating that the editorial says very little.

Since then, most likely to counter the LSA, a number of vendors have joined the Linux Standard Base project. The list includes almost all major distributions, including Caldera, Debian, Red Hat and SuSE. The Most prominent missing ones are Slackware (which uses a BSD-style boot, is quite different from the others in structure, and hence couldn't be easily standardized) and Stampede (which is growing, but is still relatively minor).

About Mike McLagan

Mike is the founder, CEO and president of the LSA. He is also the owner/president of Innovative Solutions, and the maintainer of Linux On-line. He wrote the original implementation of frame relay drivers for the Linux kernel, a routing patch for 2.1.XX kernels, and apparently, a few minor free programs (if anyone has references, please e-mail me). He has been quoted as saying that he regrets contributing to the Linux kernel, although it is unclear whether he meant it.

I also received this tidbit from Christopher Craig: ``Though it isn't of particular relevance to LSA, I think it is important to note that Mike McLagan is founder/CEO/Domain owner for Team OS/2 and is still quite active in the OS/2 community.'' Their web site does appear to be maintained by Mike's company (and it also contains advertising).

In the past, he's had a few minor run-ins with the free software community. The most prominent was about the banner ads on www.linux.org. Some people believed that a common resource, like the linux.org domain name should not be commercialized. He has stated that he has not made any profits from these, attributing the income to equipment, bandwidth and ``staffing costs''. With an estimated user-base of 7,000,000, and an estimated ten hits per user, that would come to 70,000,000 hits. At a penny an ad, that would come to $700,000 (for comparison, rates at doubleclick.net are two to six cents, depending on level of targeting). He also has three links to amazon.com from the main page (which offers referral royalties), and pushes his own businesses from linux.org. He has also consistently refused to turn linux.org over to organizations willing to maintain it without advertising. He has even been offered, in his own words, ``reasonably large sums of money'' for linux.org, and has refused to sell it (according to an e-mail I received from another source, a few thousand dollars).

Most gripes are about his strong arm tactics towards members of the free software community (see the e-mails), the intellectual property and closed membership issues of LSA, and claims that he grabs resources that there is only one of (such as the trademark 'Standard Linux' and the domain name linux.org), and tries to cash in on them. The more recent comments have focused much more on the IP and closed membership issues of the LSA, since the linux.org issue, and the forceful tactics seem to be less pertinent at the moment.

E-mails Conversations With Mike

I was able to post a few e-mail conversations people have had with Mike and other the LSA members. The first series is really worth a read. Please e-mail me if you know of any others:

Information

This is all the information I could find on the matter. Enjoy!

Standards Editorials

A number of people have written editorials on the standards situation. Freshmeat had a series of editorials from some of the more prominent figures in the free software world. Slashdot issued a call for editorials, and published two of the responses. Rob Malda sent me copies of the others, and I have e-mailed the authors asking for permission to publish the rest. A word of warning: while many of the unpublished editorials were quite good, a few were quite poorly/crudely written.

My take

Although I'm trying to present the information as objectively as possible, and let the reader draw his own conclusions, it's only fair that I tell you my own bias before you draw any conclusions. I'm opposed to the LSA, and I've lost a fair deal of respect for Mike McLagan for the way he handled this issue. I think it's fine for him to try to make a standard for the OS, even a completely proprietary standard, so long as he does not claim rights to the term ``Standard Linux.'' Claiming that his standard is the one and only is arrogant, improper and harmful to the community.

I think he should change the name from ``Standard Linux'' to something like ``LSA Standard Linux,'' move Linux On-line to a different domain, transfer Linux On-line to Linux International, and possibly send out a public apology to slashdot. If he did those three things, I would start to respect him and his contributions to the computing community quite a bit.

No major distributions are likely to adopt the LSA. Following the LSA issue, most of the GNU/Linux vendors have already jumped on the Linux Standard Base bandwagon, and LSB currently includes Red Hat, Debian, Caldera, S.u.S.E., TurboLinux, SPI, VA Research, Metro Link, Linux International and a few others. There should not be a major issue in terms of distribution support.

However, the LSA can, and is quite likely to, cause significant problems. There are still a number of issues that still need to be resolved:

These are delicate issues. I honestly don't know how to handle them. It's good to politely inform vendors joining or considering joining the LSA of the LSB, and of the inherent problems with the LSA (possibly pointing them to a page like this one). Be calm, rational, and give objective and reasonable reasons for why LSA is a Bad Idea. ``Standard Linux'' will most likely have to be fought out in court. Reclaiming linux.org for the community could only be done if we could find significant support from the key Linux kernel developers (Linus, Alan, etc.). Mike considers himself `famous,' and he does not seem to listen to people less `famous' than himself (although when I mentioned Mike to a friend who is one of the reasonably active GGI and kernel developers, my friend stated that he did not know much of/about Mike). However, `more famous' people may be able to convince him. There's a fair chance of finding support among Red Hat and Debian developers for this (although I'm not sure if that's enough; he certainly didn't consider Stampede developers important).

I'm not sure of the importance of standards for a free operating system. On one hand, standards can help push the operating system into corporate environments. On the other hand, they take away from diversity. I'm not sure how important corporate environments are to free software at this moderately early stage. Unlike commercial OSs, the success of a free operating system, is not tied to user base but to developer base. Most of the free software developers have been coding well before the user base reached millions, and will continue coding even if it started to shrink. I'm not sure if it's worthwhile sacrificing our diversity for short-term growth. I agree that we have probably been forced into developing good, free, open standards, to block proprietary ones from taking over.

As could be inferred, I view the LSA as a bad thing. I don't think a couple of minor players in the free software community would develop a competent standard for Linux. I don't think it's good that the group is closed. I think it's arrogant to grab the trademark on ``Standard Linux.''

Copyright © 1998. Piotr F. Mitros. This document may be freely modified and distributed, so long as this copyright is kept (additional names can be added, of course), and it is made clear that I am no longer the author of modified versions. If you have any questions or comments, please feel free to e-mail me at pmitros@mit.edu. I reserve the right to publish anything sent to this address.