GNU/Busybox ?!?
GNU/Busybox ?!?
Posted Mar 21, 2007 15:52 UTC (Wed) by bronson (subscriber, #4806)In reply to: GNU/Busybox ?!? by lysse
Parent article: The road to freedom in the embedded world
> We know what RMS wants
Well, apparently RMS wanted to wrest control of glibc from Ulrich Drepper, but we don't know why. At least I still don't. Drepper is very smart and a phenomental maintainer.
> Six years old, publicly available, and pretty typical of a spat between developers.
Typical? No way. Power struggles among developers tend to be rare in open source. Disagreements are resolved by code forks, not by smoky room politics. It's a wonderful way to work.
In this case, however, Stallman tried to secretly undermine Drepper. In the 10 years that I've been working with open source software, I can only think of one other situation that's even close to this: Bruce Perens trying to wrestle for control of Busybox. At least Bruce had the decency to pull his shenanigans in public.
I'm very glad to say that, unsurprisingly, both Perens and RMS failed. However, the frustration, confusion, and ill will that they caused is tangible today and will continue to sour attitudes for many years to come. It's a shameful way to work.
Lysse, can you name another case where someone has tried use his political standing to wrestle control of an open source project from an active maintainer using rhetoric and secret negotiation, not code?
Posted Mar 22, 2007 4:43 UTC (Thu)
by JoeBuck (subscriber, #2330)
[Link] (1 responses)
It is true that there was a period, years ago, when there was some fighting between Uli Drepper, RMS, and other developers, but those arguments have been mostly worked out. It wasn't just RMS vs. Drepper, the battles were more complicated than that. It just wasn't working for glibc to be Uli Drepper's personal cathedral, any more than it worked for GCC to be RMS's or Richard Kenner's personal cathedral. Hence the establishment of a steering committee, something that had worked well for egcs, something some of Drepper's admirers called a power grab by RMS.
In any case, all of glibc was and is legally assigned to the FSF, and Red Hat and Cygnus before them had blanket assignments contributing all of their employees' work. Uli would have had to quit if he really wanted to split with the FSF, and even then he couldn't change the copyright on code he already contributed; it was no longer legally his.
Posted Mar 22, 2007 6:38 UTC (Thu)
by bronson (subscriber, #4806)
[Link]
Yes, the arguments have long since been settled, but some resentment and mistrust from that situation exists to this day.
Posted Mar 22, 2007 13:07 UTC (Thu)
by lysse (guest, #3190)
[Link] (1 responses)
Is there external support for this? Has Stallman ever commented on it? Do we have anything besides Drepper's own accounts? (Note that I'm not saying Drepper misrepresented the situation; I'm sure he called it exactly as he saw it. However, when we only have half of one side it's simply not possible to get a complete objective picture of what happened, so I'd be interested to know if there's anything else out there.)
Posted Mar 22, 2007 18:21 UTC (Thu)
by bronson (subscriber, #4806)
[Link]
Here are some thread heads... I almost didn't post these because I couldn't find a single message explaining RMS's side of the situation. I despise Fox news -- I truly would like to see an alternate viewpoint. I hope someone else can find and post one?
http://www.redhat.com/archives/redhat-install-list/2001-A...
glibc was originally written mainly by Roland McGrath, while he was an FSF paid employee. It was later taken over by Uli Drepper, who has written more of it than anyone else.
your claims are incorrect
As far as I understand, Drepper was perfectly fine with the FSF maintaining copyright over glibc. He just didn't agree with RMS swooping in from nowhere and declaring mandates from on high. I do agree that RMS had some valid grievances, it's just too bad that he went about addressing them in a short-sighted, apparently politically-motivated way. Luckily RMS backed down, Drepper backed down, and sanity prevailed.your claims are incorrect
> In this case, however, Stallman tried to secretly undermine Drepper.GNU/Busybox ?!?
Well, if you're really interested, google finds a fair amount of mailing list discussion on this topic. Be warned, though: it's pretty one-sided. That might be evidence that RMS really did try to work in secret, or it might just be evidence of incomplete mailing list archives. Dunno. I didn't experience this disaster first-hand so I can only speculate.GNU/Busybox ?!?
http://sources.redhat.com/ml/libc-hacker/2000-06/msg00180...
http://slashdot.org/articles/01/08/19/2039211.shtml