rms@AI.MIT.EDU (Richard Stallman) (01/17/91)
If you have suggestions for changing the library GPL, please send them to me (rms@ai.mit.edu), not help-g++ (gnu.g++.help). If you report a problem with the way the LGPL accomplishes its aims, or a possibly unintended side effect, I'll probably do something about it. The substantial change between the first published draft and the second is due to criticisms of this sort. If you ask for a fundamental change in the aims of the LGPL, then we are unlikely to oblige. In particular, we are not considering dropping the requirement that end users have the ability and the freedom to modify the library and relink the application. True, the LGPL is a concession we are making to get more software developers to use the GNU libraries. We reluctantly decided to drop the requirement for the application code to be free, because that issue doesn't seem so close to home, and that's why we are making an LGPL in the first place. But letting a developer take away the end users' freedom to change *the GNU code* is another matter. Developers are not entitled to strip off the freedom when passing the library along. Not even when they stick it inside something else. Dropping this requirement would get some additional users for our software, but would promote freedom less. When the users saw application programs that used our library, the message they would get is that we had sold them out. The purpose of the GNU project is not maximizing the use of GNU software. It is promoting the freedom to share and change software as much as we can. Having more users is better, all else being equal, but not at the cost of forgetting what we are about. It seems that the GNU libraries will have enough use under the terms now proposed to make them thrive. Even if that's only 25% as much as they would otherwise be used, that is not a serious problem for the GNU project. Pacifism is not part of the GNU philosophy, and software hoarding is not a victimless crime. Being a wimp is not my idea of "high mindedness". 'nuff said. If people want to discuss whether they agree with the aims of the FSF, please use the newsgroup gnu.misc.discuss. It was made for that purpose.
jimad@AI.MIT.EDU (Jim ADCOCK) (01/23/91)
In article <9101161825.AA09469@mole.ai.mit.edu| rms@AI.MIT.EDU (Richard Stallman) writes: |The purpose of the GNU project is not maximizing the use of GNU |software. It is promoting the freedom to share and change software as |much as we can. Having more users is better, all else being equal, |but not at the cost of forgetting what we are about. Perhaps Stallman needs to place a "clear statement" about what GNU is about in your licensing agreements, and in your statements about "freedom" so people will not be confused by the difference between your actual goals, and some of your more casual statements. |It seems that the GNU libraries will have enough use under the terms |now proposed to make them thrive. Even if that's only 25% as much as |they would otherwise be used, that is not a serious problem for the |GNU project. You contradict yourself in two paragraphs. You state that GNU is attempting to promote the freedom to share and change software -- but not maximize the use of that software. You state that the present licenses only allow perhaps 25% of the people to use GNU-restricted software, and do not consider that a serious problem. I disagree. I believe any software offering that deliberately prohibits the majority of programmers from using it is divisive, and against the good of the software community. Programmers -- please do not allow your software to be used for such divisive ends -- please do not allow your software to be distributed under the GNU licensing restrictions. |Pacifism is not part of the GNU philosophy, and software hoarding is |not a victimless crime. Being a wimp is not my idea of "high |mindedness". 'nuff said. This again, seems to be a statement that the ends justify the means. I disagree. People use the "ends justify the means" argument to try to justify whatever set of beliefs they believe in. The end result is that other human beings are injured. Gnu software itself engages in software hoarding -- justifying that behavior in terms of trying to force others to conform to their "political" ideals. I can tolerate such behavior if it isolated. But when Stallman refers to such behavior as "freedom" I must speak out and disagree. Attempts to coerce people do not represent freedom. "Freedom" is letting people decide for themselves. The Gnu software is in no way "free" but is as restrictive if not more restrictive that software available from other more traditional companies. Please do not be deceived by Stallman's attempts to label his actions "freedom." Please do not allow Stallman's licenses to be applied to your software -- doing so does not make your software more accessible to most software programmers, but rather prevents most software programmers from using your creations. [more than standard disclaimer]
kayvan@APPLE.COM (Kayvan Sylvan) (01/30/91)
Jim ADCOCK writes: > In article <9101161825.AA09469@mole.ai.mit.edu| rms@AI.MIT.EDU (Richard Stallman) writes: > > [... too much stuff ...] > > Programmers -- please do not allow your software to be used for such > divisive ends -- please do not allow your software to be distributed > under the GNU licensing restrictions. > > [... And still more stuff ...] > > Please do not be deceived by Stallman's attempts to label his > actions "freedom." Please do not allow Stallman's licenses to be > applied to your software -- doing so does not make your software > more accessible to most software programmers, but rather prevents > most software programmers from using your creations. My apologies to the group for posting this to help-g++ rather than gnu.misc.discuss, but I couldn't help but reply to Jim's post. Jim's arguments do not really apply. The FSF licenses only restrict people's ability to restrict your access to the sources of the software you are using. You are free to do with the software what you like, as long as you don't restrict other people's freedom. Promoting freedom of software also means that you can't use free software inside your proprietary software. It is obvious that Jim from Microsoft is very invested (probably literally as well as figuratively) in the idea of restricting people's use of software for profit. He balks at the idea of not being able to use a piece of free software in a way that restricts his end-user's freedom. Too bad. To try to present that as a plea for freedom is true hypocrisy. An example that illustrates: I recently found a file system bug in the operating system I am using. Having found the cause (a buggy routine in a system library), I was able to effect a fix by replacing the buggy routine with one that worked from the net. *However*, I couldn't fix the binaries easily (because I didn't have the sources). I actually *did* end up patching the binaries (by disassembling them and findind out what to change), but it wasn't my idea of a fun time. The fact is that the software vendor had me at their mercy even though I was probably far more capable of fixing my problem in a timely fashion than they themselves were... They had, through their licensing, restricted my freedom to read and change the source. FSF software, on the other hand, is infinitely superior simply because you have the source. Making sources available to hundreds of bright intelligent software people all over the net to use/modify and send in patches insures that in a very short time you have sources that are ported to hundreds of different machines and that you have software that is *very high quality*. Enough said, ---Kayvan | Kayvan Sylvan Unix/C, GNU tools 879 Lewiston Drive 408-978-1407 | | Sylvan Associates GUIs, Databases San Jose, CA 95136 Think Globally | | Networking, X apple!satyr!kayvan Act Locally | | === Currently looking for contracts === Hire me now! Avoid the rush! === |
tiemann@eng.sun.com (Michael Tiemann) (02/14/91)
I can't stand reading Jim's disinformation any longer. I have let about 5 of his random flames fly, and this one has put me over the top. If you regularly disregard Jim's postings, you might as well disregard this one. Again, I disagree. The FSF licenses prevent the vast majority of programmers from using any libraries with such licensing restrictions. The reason being that most programmers work for companies distributing software that cannot be profitably supported in source form. When "for profit" companies see the opportunity to provide software in source form, at a reasonable support cost, then they so provide the software. A case in point being reusable object-oriented libraries written in C++. The consensus -- in the industry -- seems to be that such libraries need to be distributed in source form in order for them to be truly useful to programmers. Surprise! -- industry supplies such software in source form, without unduly restrictive licenses. This paragraph contains 3 anti-facts. First, the FSF licenses in not way restrict programmers from *using* FSF libraries. I have supported many a proprietary programming effort in my time, and they all used the FSF libraries for prototyping. Some of them even *learned* something about coding style. Second, most companies have no idea how profitable supporting source code can be; they just don't even try. I'm not saying that my software support company (Cygnus) is making Milken or Helmsley look poor, but we do alright. I'm sure that most software companies can make money in all sorts of different ways. Out of laziness, they choose to follow Microsoft's and IBM's model of software hoarding. Finally, I looked at the licensing terms of `Classix', a C++ class library. Yes, they do distribute their sources because they have to so that it works with conventional C++ compilers. But, they have the most restrictive licensing terms I have seen for a while. You cannot use, modify or distribute them in 98% of the ways I consider useful. If for-profit industry can provide source-form reusable object-oriented software libraries at hardly more than the "distribution" costs of "Free" software, and can do so without unduly restrictive licensing agreements, then, if people were to mistakenly distribute their work under Stallman's restrictive licenses, society loses. We'd all be better off if the authors of such software distributed ["sold"] the software themselves. Garbage, since I have yet to see less restrictive instead of more restrictive licenses. Also, I don't buy the argument even if true. Laws which enforce safety are a restriction on society. Does society lose by being safer, or by having the freedom to endanger itself? Should I be able to walk to my local gun shop and buy a tactical nuclear warhead or a binary nerve gas system? Maybe I should...maybe it would benefit society to keep the whining down in certain parts of Washington. Is this the kind of freedom-without-restrictions that I should enjoy? No: I prefer to compete within the restrictions of our current laws. The rest of his posting is hard to argue with, only because by now it is based on such flakey assumptions that there is no coherent point which can be argued. Suffice it to say that Jim's idea of what constitutes "overly restrictive" and what constitutes "useful" are diametrically opposed to what I have found. When I fire up a spread sheet and it doesn't do what I want, without source, I'm screwed. Since starting a company, and using software commercially, the importance I place on source availability has *increased*, as has my contempt for people who continue to argue about the unreasonableness of the FSF licenses. It's not the licenses that slow us down, it's hiring people, paying taxes, bidding too agressively, disk errors, unexpected market events, etc. In other words, by looking at our business, you can't tell that we're not just another software company, except that our customers have no problems getting source from us. Michael