Home
| Calendar
| Mail Lists
| List Archives
| Desktop SIG
| Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings Linux Cafe | Meeting Notes | Linux Links | Bling | About BLU |
Rich Braun wrote in a message to Mike Bilow: RB> Poll: who's right? Did Linus tender Linux as a contribution RB> to the GNU project, or did RMS tender the GNU environment to RB> the Linux community? ;-) GNU tools are not generally specific to Linux, and Linux was not developed with any clear long term goal in mind. I was involved with Linux when it was in the "neat toy" stage, around kernel 0.12 or so, and I can assure you that it was not especially useful. In fact, it was at that time considered a major accomplishment (and I think the phrase was "indication of maturity") that the Linux kernel could be built on a machine running Linux itself, as all early building of Linux was actually performed on Minix. GNU and FSF, on the other hand, were what we would today call "political" from the very start. The whole motivation for GNU was philosophical, not to give the world capabilities it did not have but to give it references of commonly used components which had evolved in vendor-specific ways. Even then, it was understood that fragmentation of Unix was a bad thing, and some variants were truly awful and bizarre, such as Xenix. As a result, the notion of "open source" was born partly out of necessity, and GNU even antedates POSIX. To give you some idea of how unfocused early Linux development was, I originally became involved in the development of the Adaptec SCSI driver because I was one of the few people interested who could afford a SCSI controller! (Obviously, my extensive driver-level experience with other operating systems helped, but that's not the point.) I certainly would not say that one could "run" Linux at that time, although one could experiment with it to great effect. The GNU folks did not seem to be quite sure how to respond to Linux. They dislike the open development model, which looked chaotic to them. There is no denying that the death of most such projects is an inability to get things done on time and out the door, and the GNU experience had taught them a number of organizational techniques to combat that. As a result, the GNU sense seemed to be that Linux would eventually come unraveled, and that a closed development model such as used by FreeBSD would have a better chance of survival. FSF also had its own operating system project, what we now know as "Hurd," which was treated similarly to the Holy Grail, in that there is a lot of seeking and not much finding. So my perception was that the FSF view of Linux was at first honest pessimism mixed with dismay, especially since at that time no one expected Linux to become platform-independent and run on anything other than the Intel architecture. Eventually, the FSF decided to try to fight this sort of chaos directly by joining forces with the Debian project, which was intended to provide an organizational infrastructure to stand behind an officially endorsed FSF distribution of Linux, very much the way Cygnus stands behind GCC or Aladdin stands behind Ghostscript. Ironically, although Debian turned out to be a great success on exactly these original terms, things started badly and led to a parting of the ways with FSF. It was the easy availability of GNU tools, especially the C compiler, which made Linux possible. Before Linux was even imagined, GCC was a mature product and the most important of the GNU line. Emacs, for example, played no important role in the existence of Linux, but no one would have even started on Linux if developing a C compiler was going to be the first step. Linus himself has often joked that many of his decisions have been motivated by laziness. That is a serious joke, since laziness leads to efficiency. The greatest single GNU contribution to Linux, of course, is the GPL. Besides that, however, I can't see regarding Linux as in any way a GNU project. The fact that Linus had no long term goal for his project is made clear by his failure to name it, and "Linux" was coined by other people to tweak Linus much in the same spirit with which "Unix" was coined to tweak "Multics." -- Mike - Subcription/unsubscription/info requests: send e-mail with "subscribe", "unsubscribe", or "info" on the first line of the message body to discuss-request at blu.org (Subject line is ignored).
BLU is a member of BostonUserGroups | |
We also thank MIT for the use of their facilities. |