-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 At some point hitherto, Rob Ransbottom hath spake thusly: > On Wed, 28 Aug 2002, Derek D. Martin wrote: > > > My own experience has largely been the opposite. I think it's more a > > question of what you're more familiar with and more comfortable with, > > than which is actually better. > > I agree, familiarity is the primary issue. Debian does have a better > track record but at the cost of being slower to release. Even there, I think it depends on what you want to use it for. I've come across bugs in Debian which absolutely prevented me from using it. Their sendmail package, for instance, seems to work differently from sendmail everywhere else in the world, to the point where getting it to do certain kinds of address rewriting is either impossible or extremely painful. The response I usually get from debian users to this complaint is "use postfix." Which is fine; postfix works great. But it's not a very good answer if I WANT to use sendmail. The point being, Debian is better for Debian users. Red Hat is better for Red Hat users. If that were not true, they would switch to something else. I've been using Linux since 1995, have used both Debian and Red Hat (and Mandrake and Slackware and...), and still favor Red Hat. It does what I want it to. And as far as fixing reported bugs goes, in my experience Red Hat does it faster than Debian at least as often as Debian does it faster than Red Hat. > > Red Hat Linux is not GNU/Linux, it's Red Hat Linux. There is only one > > GNU/Linux: debian GNU/Linux. None of the rest of the Linux > > distributions have caved to rms's requests based on entirely spurious > > arguments to call their Linux distributions "GNU/Linux;" and as > > I don't doubt rms' arguments were spurious, that does not de-legitimize > the request itself. How does it not? The vendors name their products; that is their right. Even in the world of printed text, the producer of a compilation names the compilation; major contributors only have a say in the naming of such a compilation at the whim of the publisher. Barring legal basis to the contrary (such as by contractual agreement), rms has absolutely no right to compel anyone to use another name than that which the vendors provide. > Redhat/Mandrake/SUSE may be as much GNU/Linux as Debian but > naturally they are less willing to advertise the fact. What exactly makes them GNU/Linux? The fact that they include a fair amount of (OPTIONAL) GNU application software that may or may not ever be installed by the user? It's true that there is a large amount of GNU software shipped with a Linux system; almost none of which is actually required to be installed for the system to be useful. This is especially so to the average computer user, who will live almost entirely in a GUI environment, reading e-mail, browsing the web, and using office suite software. The only GNU software which is required for that to happen on a linux system is glibc (and maybe a few small supporting libraries). Period. It may even be true that all of that software was compiled with gcc, but if I were to use a Windows system with a MyCo web browser and MyCo office suite that was compiled by Borland C/C++ compiler, does that make it Borland/Windows? Of course it does not. Having your application software shipped with a system is hardly a reason to usurp the name of the thing. It doesn't even give Borland the right to call the office suite Borland/MyCo Office. IMO, the concept of a GNU/Linux system is a fallacy; there is no such thing (excepting Debian, see below). The bottom line is Stallman wants credit; he wants the spotlight. His ego was bruised when the focus was on Linus Torvalds, and his GNU/Linux campaign is his selfish attempt to heal it. Frankly, I find it repugnant, and in some ways directly counter to some of the ideas that he is trying to publicize in so doing. Part of the reason that Linus has remained in the spotlight for so long, IMO, is precisely because he was always the timid but likable geek who was much more interested in writing good code for people to use, than in getting the credit for it... > I'm not a 'free software' partisan. I have never bought > rms' "information wants to be shared" tripe. I AM a free software partisan, and support many of the philosophies espoused by rms, albeit to a lesser degree. Nevertheless, his attempts to rename other people's products are misguided. I do not let my respect for the man or his work cloud my judgement as to what is right or fair. There are much better ways for rms to go about trying to get the credit that he deserves (which I agree is substantial) without offending people. > It didn't seem like a cave-in when Debian added GNU. It seemed like > a recognition of where many key and enabling elements of Debian came > from. And it is their right to make that recognition. But it is also every other vendor's right NOT to make that recognition. > I only saw this as a Debian user, it seemed like the blip you'd expect > it to be. I just figured it won a vote. I didn't care for the name, > didn't see it as necessary, but could see it as reasonable, by the > reasonable man test. If you have juicy insider gossip, share it. In the case of Debian, it does actually make a certain amount of sense. The FSF supported the creation and growth of Debian directly. That's not true of any of the other distributions. For them, it was not a cave-in; they were simply giving the GNU Project and the FSF the recognition that they felt (and I think rightly so) they deserved. http://www.debian.org/doc/FAQ/ch-basic_defs.html#s-gnu http://www.debian.org/doc/manuals/project-history/ch1.html#s1.1 > Don't feel bad, crack heads are far above those gutter scum, > bottom quoters. - -- Derek Martin ddm@pizzashack.org - --------------------------------------------- I prefer mail encrypted with PGP/GPG! GnuPG Key ID: 0x81CFE75D Retrieve my public key at http://pgp.mit.edu Learn more about it at http://www.gnupg.org -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iD8DBQE9blqMdjdlQoHP510RAklhAKCDQENdniF8soXbTYpZu72FrGh1QACePVM7 4iRteM9hSjS+/3Ao/0cYgnc= =RA7x -----END PGP SIGNATURE-----