The GNU C++ compiler. He is still active in development (or was recently), he implemented pre-compiled header file support for GNUPro, the Cygnus version of GCC.
Uh, the Marshall plan alone (the only thing I associated with the man before reading Brins article) would qualify him to win an European "man of the century" award. The Marshall plan wasn't just about giving away money, it was given in forms that forced the European nations to cooperate, in order to prevent a new war. In a way, George Marshall founded the European Union.
In my experience, free software is often not first to market.
Maybe so, but your examples doesn't support it.
It took Emacs decades to mature.
Huh? What makes the current Emacs versions "mature" and the previous versions not? Emacs has been _the_ most complete text editor as far back anyone remembers. When a new major version come out, it is often somewhat unstable for the first few versions. But that is cyclic phenomen, it makes little sense to call Emacs 20.5 for more "mature" than Emacs 18.55.
And GNU C++ didn't become a reasonably complete C++ compiler years after good commercial implementations were already out there.
The various C++ compilers, including g++, have leapfrogged each other in terms of usability. There is no clear pattern, except that Stroustrups own CFRONT of course was first.
Mature C/C++-based GUI toolkits took a few years to come out after mature commercial C/C++ GUI toolkits.
Wrong again. InterViews was one of the first, and it was free. I wonder what "mature" C++ GUI toolkit you are thinking of.
Same for free, mature, powerful UNIX-like kernels.
Well, Unix was proprietary from day one. The first reimplementation was BSD, and it ended up free.
Ask him if you should then back Opera over MSIE and Netscape Navigator, since it is the most sold (rather than downloaded for free) browser, and thus by his logic the "market leader".
I find it amazing that after all the anger and hostility you've shown towards GNU and the Free Software Foundation, you have summed up the GPL and blessed it with your stamp of approval as a 'true free software license'.
You are assuming TC's hatred has a rational basis. This is a natural assumption, given that TC is quite well formulated, and explain why his flames are moderated up. However, TC's has quite a different worldview. Basically, anything Perl is good, anything that in any way can be seen as competing or critisizing Perl is bad. RMS became "bad" when he complained about the lack of good, free introductory texts about Perl. Since that time, TC has campaigned against anything related to RMS, FSF or GNU. The GPL isn't evil because of anything it says, it is evil because it was written by RMS.
GPL doesn't prevent forks, in fact, the right to fork a project is something the GPL was designed to protect.
However, since all forks must be GPL'ed as well, it will always be legal to remerge the forks. This is a big difference compared to licenses that allows proprietary forks, for example the BSDL. The many proprietary BSD derived operating systems can not be remerged without permission of all the copyright holders.
How's that again? Emacs is not an open source project? It does not invite patches from everybody? It does not incorporate contributions from an army of individual hackers? It has not made all its bugs shallow by offering its source code to millions of eyeballs?
Emacs development is very closed compared to e.g. the Linux kernel. The pre-tester lists are invitation only, and doesn't contain discussion. Test releases are rare, and placed in execute-only directories on secret ftp sites. There are, to my knowledge, no real developer lists, instead RMS send private mail to people whose help or input he want on particular issues. There are no access to the development code outside an even smaller group than the pre-testers.
Contrast this with XEmacs development, where the development discussion is public, there are frequent development releases, and anonymous CVS access to the latest sources.
ESR has been involved with Emacs development, and his characterization is quite on the mark. The Emacs distribution itself is a one man project, however, a community (or bazaar) exsists around the various Emacs Lisp packages.
The GPL doesn't create conflicts. It demonstrates, like all licenses, that conflicts of interest exists. If there were no conflicts of interest, there would be no need for licenses.
The conflict of interest that justify the GPL, is the interest of (some) free software programmers to ensure that all public enhancements to their code remain free, as opposed to some companies interest in keeping their enhancements to the code unfree.
One way to resolve this conflict would be, as you suggest, for the free software programmers to give up their right, and let these companies keep their enhancements unfree. This particular brand of conflict resolution is called "unconditional surender."
Another way to solve the conflict would be to note that the interest of the free software programmers are in no way helped by the unfree enhancements, while making free enhancements have proven valuable to many companies. This leads to to potential conflict resolutions: One is that the company ignore the free software, in which neither part have lost anything. The other is that the company makes free enhancements to the code, in which case both parts win.
This later way of resolving the conflict is, to me, much prefereable than your suggestion which basically amounts to telling the free software programmers to "bend over and let them bugger you".
RMS is drafting a "free book" license
on
Free Books Online
·
· Score: 2
So I assume he also feel the GPL is inadequate for books. Also, note that the books published by the FSF aren't under the GPL either, but under ad-hoc licenses.
There are ports of the X libraries for cygwin, and there are gratis X servers that run under NT.
I haven't bothered with them, as Cygwin 1.0 is close to a perfect fit for my needs. It gives me a nice Unix programming environment under NT for producing win32 executables, and since XEmacs runs natively under win32 (and is included on the CD from Cygnus), I cannot think of any X11 programs I miss.
There was such a project, basically it was to be a Debian format distribution of Cygwin. It would have been very useful, given that Cygnus doesn't put in much work on making free distributions, but nothing seems to have come from it. Nobody volunteered to do actual work.
It could be a lot of fun if you e.g. could communicate with the computer through gestures. It would probably not do much to spreadsheets, but gaming and chatting might benefit.
Yeah, that's the media angle, and probably why some lusers are attracted to Linux. However, I doubt it is the motivation for a significant fraction of the developers or contributers.
Not being able to fix a bug in an application you use and share the fix is so annoying for a programmer, once you have tried using free software.
Loser> Good ol' vi. Let's see EMACS do this with as few key strokes.
Heh! Line numbers is a brain damage inflicted by harmful technology like VI and BASIC. With Emacs, you simply select the text you want commented in or out.
-frepo works fine for me on Solaris, but not on any other platforms where I have tried it (aix, hp/ux and cygwin). The official policy of the gcc maintainers is that -frepo is depreciated, and linker smartness is the way to go.
Precompiled were implemented by Michael Thiemann at Cygnus for Corel, but apparently the gcc maintainers doesn't like the implementation, so it may never reach the public distribution.
> Wanna bet it's *all* copyrighted by the FSF? I > submitted a small patch once, and it was > accepted, and I never signed any forms.
The policy is that contributions of less than 10 lines of code are too small to fall under copyright law, and thus does not need papers. I don't know US law in details, but that would certainly be true under Danish law. Also, sometimes the maintainers rewrite patches rather than ask for papers.
Maybe his conclusions are correct, but in that case they definitely aren't based on his understanding of history, which is non-existent.
1. Unix.
Basically ok, as a kids version of history.
2. BSD
MachTen, NeXTStep, SunOS belong to the "Unix" camp above. OpenBSD split from NetBSD, and can thus not be characterized as a splinterproject from 386bsd.
3. Emacs
Totally crap. GNU Emacs was always under a "remain free" license. The non-free Emacsen were all written from scratch. Lucid Emacs was not only free, but the code ownership was assigned to the FSF for merging back. However, later Sun contributed code which was still GPL, but *not* assigned to the FSF.
The real story is that RMS wanted to keep total control over Emacs development, and refused to release Emacs 19 when it suited Lucid commercial interests. Today a merge is prevented partly because of the control issue, and partly because of the unassigned Sun code the FSF don't want to use.
4. httpd
I never followed that one.
5. gcc
Crap. *intel* made a pentium optimized port of gcc, and released it as GPL. They did *not* assign the code to the FSF. The version released by Intel did *not* work on any other platforms. So there were never any possibility of a remerge. Intel has later payed Cygnus for developing a new Pentium II optimized backend, which *will* be in gcc 3.0.
Egcs was created because gcc 2.8 never seemed to materialize, and in particular the C++ frontend of gcc 2.7.2 was embarrassingly old and buggy compared compared to what the Cygnus engineers had developed. Also, the Cygnus engineers found it very hard to attract outside developers with the closed developing model of gcc. Egcs was created as an experiment, with RMS blessing, to demonstrate the efficiency of a more open development model. It was a success.
6. glibc
Ok, except that the split then probably wasn't a mistake. Linux needed a working libc *then*, they couldn't wait until glibc was finished.
Being lazy has nothing to do with it. It is a question of not being stupid.
Evidence pointing toward your lies containing any truth: An anonymous posting on a web board.
Evidence pointing towards your lies being totally made up for the occation: All prior experience with the FSF and RMS.
There really aren't no comparison. Making up anonymous claims on web boards with no evidence is no effort, putting in active effort to disprove them would be a losing strategy.
However, iff someone here are inclined to put any value in what they read from anonymous sources on/., check out the signed contracts gcc contributers gets from the FSF. It should be relatively simple to show that such "special closed source licenses" would not be legal.
Now, I happen to use XEmacs. It's my favorite editor. I couldn't code without it, debug without it, or even read e-mail without it. But I can browse the web without it, and I think building an emacs-based browser is just way over the edge.
But W3 is what allows Gnus to read/.. Why would anyone use a web browser to read a message board, when a news reader is so much more convenient? Also, W3 allows Gnus to read mail (and news) messages in text/html format. It's so fast and tranaparent that I often forget to be angry over getting text/html. In fact, I rarely notice it.
Slashdot Mirror
The GNU C++ compiler. He is still active in development (or was recently), he implemented pre-compiled header file support for GNUPro, the Cygnus version of GCC.
Unfortunetely, the one thing vi does well is text maculation. So it doesn't qualify in the text editor category.
Uh, the Marshall plan alone (the only thing I associated with the man before reading Brins article) would qualify him to win an European "man of the century" award. The Marshall plan wasn't just about giving away money, it was given in forms that forced the European nations to cooperate, in order to prevent a new war. In a way, George Marshall founded the European Union.
Ask him if you should then back Opera over MSIE and Netscape Navigator, since it is the most sold (rather than downloaded for free) browser, and thus by his logic the "market leader".
Of course, Emacs does do a lot more than Mozilla.
You are assuming TC's hatred has a rational basis. This is a natural assumption, given that TC is quite well formulated, and explain why his flames are moderated up. However, TC's has quite a different worldview. Basically, anything Perl is good, anything that in any way can be seen as competing or critisizing Perl is bad. RMS became "bad" when he complained about the lack of good, free introductory texts about Perl. Since that time, TC has campaigned against anything related to RMS, FSF or GNU. The GPL isn't evil because of anything it says, it is evil because it was written by RMS.
GPL doesn't prevent forks, in fact, the right to fork a project is something the GPL was designed to protect.
However, since all forks must be GPL'ed as well, it will always be legal to remerge the forks. This is a big difference compared to licenses that allows proprietary forks, for example the BSDL. The many proprietary BSD derived operating systems can not be remerged without permission of all the copyright holders.
It is still free and now also protected by the GPL.
Larry Wall created patch as free software. After he lost interest in it, the FSF took over maintenance.
Emacs development is very closed compared to e.g. the Linux kernel. The pre-tester lists are invitation only, and doesn't contain discussion. Test releases are rare, and placed in execute-only directories on secret ftp sites. There are, to my knowledge, no real developer lists, instead RMS send private mail to people whose help or input he want on particular issues. There are no access to the development code outside an even smaller group than the pre-testers.
Contrast this with XEmacs development, where the development discussion is public, there are frequent development releases, and anonymous CVS access to the latest sources.
ESR has been involved with Emacs development, and his characterization is quite on the mark. The Emacs distribution itself is a one man project, however, a community (or bazaar) exsists around the various Emacs Lisp packages.
The conflict of interest that justify the GPL, is the interest of (some) free software programmers to ensure that all public enhancements to their code remain free, as opposed to some companies interest in keeping their enhancements to the code unfree.
One way to resolve this conflict would be, as you suggest, for the free software programmers to give up their right, and let these companies keep their enhancements unfree. This particular brand of conflict resolution is called "unconditional surender."
Another way to solve the conflict would be to note that the interest of the free software programmers are in no way helped by the unfree enhancements, while making free enhancements have proven valuable to many companies. This leads to to potential conflict resolutions: One is that the company ignore the free software, in which neither part have lost anything. The other is that the company makes free enhancements to the code, in which case both parts win.
This later way of resolving the conflict is, to me, much prefereable than your suggestion which basically amounts to telling the free software programmers to "bend over and let them bugger you".
So I assume he also feel the GPL is inadequate for books. Also, note that the books published by the FSF aren't under the GPL either, but under ad-hoc licenses.
There are ports of the X libraries for cygwin, and there are gratis X servers that run under NT.
I haven't bothered with them, as Cygwin 1.0 is close to a perfect fit for my needs. It gives me a nice Unix programming environment under NT for producing win32 executables, and since XEmacs runs natively under win32 (and is included on the CD from Cygnus), I cannot think of any X11 programs I miss.
> Debian GNU/NT, anyone? ;-)
There was such a project, basically it was to be a Debian format distribution of Cygwin. It would have been very useful, given that Cygnus doesn't put in much work on making free distributions, but nothing seems to have come from it. Nobody volunteered to do actual work.
It could be a lot of fun if you e.g. could communicate with the computer through gestures. It would probably not do much to spreadsheets, but gaming and chatting might benefit.
Not being able to fix a bug in an application you use and share the fix is so annoying for a programmer, once you have tried using free software.
eVIl> x,ys/^/\/\//g
Emacs> C-c C-c
eVIl> x,ys/^\/\///g
Emacs> C-u C-c C-c
Loser> Good ol' vi. Let's see EMACS do this with as few key strokes.
Heh! Line numbers is a brain damage inflicted by harmful technology like VI and BASIC. With Emacs, you simply select the text you want commented in or out.
-frepo works fine for me on Solaris, but not on any other platforms where I have tried it (aix, hp/ux and cygwin). The official policy of the gcc maintainers is that -frepo is depreciated, and linker smartness is the way to go.
Precompiled were implemented by Michael Thiemann at Cygnus for Corel, but apparently the gcc maintainers doesn't like the implementation, so it may never reach the public distribution.
> Wanna bet it's *all* copyrighted by the FSF? I
> submitted a small patch once, and it was
> accepted, and I never signed any forms.
The policy is that contributions of less than 10 lines of code are too small to fall under copyright law, and thus does not need papers. I don't know US law in details, but that would certainly be true under Danish law. Also, sometimes the maintainers rewrite patches rather than ask for papers.
Maybe his conclusions are correct, but in that case they definitely aren't based on his understanding of history, which is non-existent.
1. Unix.
Basically ok, as a kids version of history.
2. BSD
MachTen, NeXTStep, SunOS belong to the "Unix" camp above. OpenBSD split from NetBSD, and can thus not be characterized as a splinterproject from 386bsd.
3. Emacs
Totally crap. GNU Emacs was always under a "remain free" license. The non-free Emacsen were all written from scratch. Lucid Emacs was not only free, but the code ownership was assigned to the FSF for merging back. However, later Sun contributed code which was still GPL, but *not* assigned to the FSF.
The real story is that RMS wanted to keep total control over Emacs development, and refused to release Emacs 19 when it suited Lucid commercial interests. Today a merge is prevented partly because of the control issue, and partly because of the unassigned Sun code the FSF don't want to use.
4. httpd
I never followed that one.
5. gcc
Crap. *intel* made a pentium optimized port of gcc, and released it as GPL. They did *not* assign the code to the FSF. The version released by Intel did *not* work on any other platforms. So there were never any possibility of a remerge. Intel has later payed Cygnus for developing a new Pentium II optimized backend, which *will* be in gcc 3.0.
Egcs was created because gcc 2.8 never seemed to materialize, and in particular the C++ frontend of gcc 2.7.2 was embarrassingly old and buggy compared compared to what the Cygnus engineers had developed. Also, the Cygnus engineers found it very hard to attract outside developers with the closed developing model of gcc. Egcs was created as an experiment, with RMS blessing, to demonstrate the efficiency of a more open development model. It was a success.
6. glibc
Ok, except that the split then probably wasn't a mistake. Linux needed a working libc *then*, they couldn't wait until glibc was finished.
Being lazy has nothing to do with it. It is a question of not being stupid.
/., check out the signed contracts gcc contributers gets from the FSF. It should be relatively simple to show that such "special closed source licenses" would not be legal.
Evidence pointing toward your lies containing any truth: An anonymous posting on a web board.
Evidence pointing towards your lies being totally made up for the occation: All prior experience with the FSF and RMS.
There really aren't no comparison. Making up anonymous claims on web boards with no evidence is no effort, putting in active effort to disprove them would be a losing strategy.
However, iff someone here are inclined to put any value in what they read from anonymous sources on