On Wed, May 29, 2002, Matan Ziv-Av wrote about "Re: Elections": > You completely miss RMS' point. Did you ever read his explanation > (http://www.gnu.org/gnu/linux-and-gnu.html), or only what his oponents > claim his reason is? (or the slashdot version).
I read him, and understand him completely. It doesn't mean I always agree with him. > This has nothing to do with specific programs, but has to do with the > system. Unlike Linus, who decided to write a kernel, Knuth, who wrote a > typesetting system, etc. the gnu project was about creating a free > _system_. They had no problem using free software they did not write, > and they wrote whatever pieces were missing. This line of reasoning is bogus; If I announce "Nadavix" and announce that "I have no problems using free software I did not write", would it make me justified in demanding that other people call their systems Nadavix too? Of course not, it is the software that I *did* write, and its weight in the resulting system, that might give me such a justification. I should be calling the system I'm working on now (Redhat 7.3, by the way) "GNU" only if GNU is indeed the major part of this system. What the people who disagree with Stallman in this issue, and I partially agree with them, are saying is that GNU is not that much more important than other components of this system (e.g., X11), at least not enough to give them the right to demand the system be named after them. If it was some sort of tradition (like the name "Linux" is), it would have been different, but it is *not* tradition to say "GNU/Linux", and it is quite silly to start one now. Unlike other people who wrote in this thread, I'm not *against* calling the system GNU/Linux (apart from such a name being long and ugly), if that was the common tradition. But you have to face reality - it isn't. The name Linux has stuck so hard in people's minds, that I hear (as I said earlier) people calling "Linux" to commercial Unix, and even people asking me if I have "Linux 7.3" (when they obviously mean Redhat 7.3). > When any linux distributor started to create a distribution (RH, > Yggdrassil, SLS, etc.) they did not start from Linux kernel and started > adding programs - they started with gnu system + Linux kernel and added > programs. Personally, I switched to Linux (from AT&T System 5 release 4) when it got X11, something which wasn't working at that time on my SVR4 operating system. So maybe X11, not GNU, should get the credit for the success of "Linux"? Again, I'm not claiming that GNU are not important - perhaps they are even the single most important part of the Redhat system I'm working on now. But they are not the only important part and therefore don't get (in my opinion) priority in their demands that "GNU" is stuck as part of this system's name. This is a practical issue, don't forget it: the name of the system has to be kept short, and is NOT the place for a list of acknowledgments. > > etc.) are becoming more and more important. A person might use a "Linux" > > machine rarely using GNU utilities: he might be mostly using KDE graphical > > tools to browse directories, files, and so on, and when he does need a > > shell he might be logged in through openssh (non-GNU), using Zsh (non-GNU), > > editing with Vim (non-GNU), programming in Perl (non-GNU), compressing with > > bzip2 (non-GNU), etc. You get the picture. > > Of course I get the picture. If we ignore gcc, glibc, binutils, fileutils, > bash, shellutils, texutils, gnome, autoconf, emacs, gdb, ghostscript, > less, groff, then we see that we don't use any software written by gnu) Look, I agreed that gcc, glibc and binutils are very important GNU projects that don't have any real substitutes now. Ok. But don't tell me that things like fileutils, shellutils, textutils, are what lets you name a system - for all I care Redhat could switch to the BSD versions of these utilities instead of the GNU ones (like it chose to include slocate instead of GNU's original locate). Some people (not me) would even say "let's drop all these completely and use only KDE GUIs". Bash, Emacs, Less, Groff, Gnome - these are all things that are nice, useful and important, but that have non-GNU alternatives, some even better alternatives. The Redhat system I am using has these alternatives preinstalled, and I could be (and in fact am) using these alternatives more than the GNU tools. So again I don't see how exactly this explains why GNU gets the first-rights to name the system. P.S. Imagine Microsoft insisted that every software solution based on Microsoft Windows will be called Microsoft/Gizmo. Obviously, they don't. When you let other people use your software you keep certain rights, but one right you certainly don't keep is the right to make demands on how your integrators name their final product. If anything, Microsoft would insist that that you do NOT call your gizmo Microsoft/Gizmo, because that would dilute their trademark, cause people to blame them for Gizmo's bugs (as if they don't have enough of their own), and so on. I think Stallman hasn't thought this problem through. What if I create a really crappy and buggy Linux distribution, and name it Nadavix GNU/Linux? People might start associating bad quality with the name "GNU" instead of the name "Nadavix", where the blame rightly belongs. Is that what he wants? -- Nadav Har'El | Wednesday, May 29 2002, 19 Sivan 5762 [EMAIL PROTECTED] |----------------------------------------- Phone: +972-53-245868, ICQ 13349191 |Ways to Relieve Stress #10: Make up a http://nadav.harel.org.il |language and ask people for directions. ================================================================= To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
