To Whom It May Concern:

I am computer professional and I would like to contribute my comments on the proposed software warranty legislation. I am a strong supporter of the Open Source movement as well as the Free Software movement.

Allow me to begin by stating that in certain circumstances I support mandating that software companies provide responsible warranties for their software products. In that, I mean that if a large (public or private) corporation were to release a software product and spend millions of dollars on sales and the marketing of the product, then that corporation should have an appropriate warranty to maintain liability if the product does not meet certain quality guidelines. Granted, the market does eliminate many poor-quality products, but in the case of poor software, many consumers are left holding the shoddy product and unable to return it. Has anyone ever tried to return a fully purchased (not OEM) copy of the Windows operating system? Part of the loophole that companies like Microsoft use are software patches. Microsoft will not refund money, but will [too] often issue software patches for all of their products. Usually, getting the software patch released to the public is not an overnight affair, so consumers are forced to either turn off their computers or disable certain services that make the machine functional in the first place while waiting up to several months for a patch. The patch, however, is not always the end of the problem. In the case of Microsoft, a patch will solve one problem, but create a handfull more. The fact that the Windows source code is closed off means that anyone who uses Windows must live will all its problems and computer developers must depend on Microsoft to fix everything.

In the case of Open Source and Free software, the rules are different. Open Souce and Free software are subject to intense peer review, much like science in general. If a scientist said that millions of Pamela Anderson look-alikes were living on the surface of the moon, the rest of the scientific community could (and would!) easily prove him wrong. If such a scientist made too many such false claims, he would lose credibility with the community. This peer review alone is a very strong system by which to assess the quality of software. Open Source software means that the programming source code is open for anyone to read and modify. The advantage of this is that if I purchased a software product that was Open Source, and something goes wrong, I can look at the code fix it myself if the company will not be responsible. I could also read comments and reviews from programmers who are stronger experts than I, and obtain information on any potential problems before I buy. Free software is not always Open Source. However, the price makes putting a warranty on it useless.

Compare this to the price of used cars. I could buy a very inexpensive 1970 model car, but there would be no warranty--it would be sold as is. I have downloaded numerous peices of free software for both Windows and Linux. In the free Windows software, I usually find that the author has inserted code in his program to run services in the background without my knowledge. For example, I downloaded a Windows program to un-zip files. I later found that this program had been running another program to track my movement on the internet and relay that information back to its author. While the same thing could be done under Linux, it is much less likely to succeed on a properly configured Linux system. Most programmers who write free software for Linux don't insert malicious code into their programs to begin with. Problems can still arise when installing new programs, though. Linux is such a stable operating system that users often to not do an entire system upgrade every few months. They simply upgrade whichever programs have bugs, but leave the rest alone. Given that, a user may be running an old version of the Linux kernel (the core of the operating system). It is very possible that the user could install new software (free or not) and have problems due to the older kernel. Many times, though, this exposes bugs in the installed software and/or the kernel itself. Usually such bugs are remedied within a few hours--because of Open Source and peer review.

To force cash cow companies like Microsoft to stand by more responsible warranties would be a blessing (be careful, though--Microsoft is notorious for extra fine print giving it loopholes!). However, to force the Open Source and Free Software communities to use the same kinds of warranties would be a mistake. Many of the programmers in these movements are volunteers from around the world, so there is a jurisdiction issue. Aside from that, though, programmers in these movements are responsible enough not to have such behavioral enforcement. The peer review system is not in the hands of an elite minority of experts--it is in the hands of an enourmous group of programmers who all share the anonymity of the internet. In these movements, good software flourishes and bad software is ignored. Please let these movements remain as they are. If you plan to still proceed with more regulation, then please be as precise as possible in the wording so all affected programmers will clearly understand it.

Sincerely,
Charles Mire
Systems Analyst, EDI
Dr Pepper/Seven Up