Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!iuvax!cica!tut.cis.ohio-state.edu!purdue!gatech!udel!princeton!phoenix!markv From: markv@phoenix.Princeton.EDU (Mark T Vandewettering) Newsgroups: comp.misc Subject: The GNU Public License Message-ID: <9655@phoenix.Princeton.EDU> Date: 29 Jul 89 16:22:43 GMT References: <26@ark1.nswc.navy.mil> <26719@agate.BERKELEY.EDU> <5351@ficc.uu.net> Reply-To: markv@phoenix.Princeton.EDU (Mark T Vandewettering) Organization: Princeton University, NJ Lines: 84 People have been going back and forth with Peter De Silva about the GNU public license, claiming that he doesn't understand or appreciate the goals of the FSF and Richard Stallman. Could quite conceivably be true, I don't understand or appreciate them especially either. I have suffered a bit of disillusionment when it comes to GNU software. Originally, I thought the goals were to create a set of Unix-like tools that we could all freely use, without the restrictions placed on AT&T's implementation of Unix. We would be free to modify, change, improve and POST source to our improvements, without the restriction of copyrights and licensing agreements. Unfortunately, this is NOT the goal of GNU-ware or the Free Software Foundation. The FSF seeks to establish a different kind of property right, by claiming rights to any software which "uses" GNU software, where "uses" varies from program incarnation to program incarnation. As an example, consider the case of Bison versus that of GCC. Anything compiled with GCC is not covered by the GNU copyleft (although RMS has admitted that that would be a "right" thing to do). Anything made using bison integrates a "skel" file, so it is considered a derivative work. Excuse me, but there seems to be a very VERY subtle difference between these two instances. The assembly output of a compiler is in some sense a "derivative" of the compiler. Its true that it doesn't integrate the text of some file into its output, but the result of its output is the derivative of the compiled rules for parsing and code generation. With the proper perspective, the differences between the two licenses seems almost nonsensical. The true fact of the matter is, the only real reason why RMS and the FSF grant you the right to your own binaries is the fact that he really can't enforce taking them away and granting them to the FSF. People have claimed that the main reason the GNU copyright exists was to ensure the rights to give your code away. Heck folks, you always have that right, I give my code away all the time. Nobody can ever keep you from giving your code away. (Before people contradict by saying you can't give code that you right at work away, it isn't yours to begin with, so of course you can't.) The other claim is that the GNU project produces better programs than any commercial or public domain scheme, because it allows the free redistribution, and demands the redistribution of improvements. This may be true, I believe RMS to be the single most gifted programmer I have ever seen, he rights excellent code and is phenomenal in his output. But on the other hand, consider where GNU ware would be without the unthinkably large body of users and programmers who serve as beta-testers for the GNU project. Sure, we have a stake in it, our improvements will get fed back to us in the next release, which usually translates to once a month for gcc, less often now for emacs. But the only reason we have this feedback loop is because of the wide distribution of GNU ware over the Internet. Without a high bandwidth, large audience presence on the Internet, GNU's progress would have been much slower than it is. The bottom line: if you want to make a living out of programming, you or your company needs to copyright your code in order to maintain your rights to sell the code. You have a perfect right to do so, because you have incurred an economic penalty (read: living expenses) in order to right this program. If you can get someone to pay money for your program, then you have successfully entered the world of capitalism. If on the other hand, you like to program, you like to have your work used by others, you like to see their improvements, or you just like to hear "Thanks" occasionally, you can release your program into the Public Domain. You have relinquished commerical interest in your program, and can enjoy the fact that people will use your program and improve it. The GNU project is neither of these: it seeks to deny your rights to make a living from software which they "give" to you. The license is confusing and misleading, as is the idealism with which it is usually argued on the net. The pro-GNU forces will undoubtably flame me soundly, that's okay, I have my asbestos undies on. They seem to outnumber the counter-GNU forces by a considerable margin. That's okay too, I am often in the minority. (Let the flames begin) Mark T. VandeWettering