Having read about the libc5 fork of glibc in Rick Moen's "Fear of Forking" essay, I went looking for more info about this fork since it seems to have healed so well that no one ever talks about it anymore. It also seems strangely underdocumented for a 10 year old event in a community that values putting everything about everything online.
http://fsfe.org/en/fellows/ciaran/ciaran_s_free_software_notes/history_of_gl...
If anyone knows any other sources of info that I didn't reference, I'd be happy to hear about them.
On Tue, 2007-06-19 at 14:24 +0100, Ciaran O'Riordan wrote:
Having read about the libc5 fork of glibc in Rick Moen's "Fear of Forking" essay, I went looking for more info about this fork since it seems to have healed so well that no one ever talks about it anymore.
I still have machines which run it :)
You might find this interview interesting:
http://linuxgazette.net/issue32/tag_libc5.html
I remember another one, but can't find it I'm afraid.
Cheers,
Alex.
Alex Hudson home@alexhudson.com writes:
Interesting. I've added it now.
I remember another one, but can't find it I'm afraid.
Whenever you remember, let me know and I'll add it.
On Tue, Jun 19, 2007 at 02:24:13PM +0100, Ciaran O'Riordan wrote:
Having read about the libc5 fork of glibc in Rick Moen's "Fear of Forking" essay, I went looking for more info about this fork since it seems to have healed so well that no one ever talks about it anymore. It also seems strangely underdocumented for a 10 year old event in a community that values putting everything about everything online.
http://fsfe.org/en/fellows/ciaran/ciaran_s_free_software_notes/history_of_gl...
As you write "just before my time". My opinion is, that because of that your problem is.
If anyone knows any other sources of info that I didn't reference, I'd be happy to hear about them.
Let me express it (a wild guess) and save your time for more useful things.
What Linux was in 1994/1998? I bet you have no idea. But some major points can be derived by deduction. Just a few of them follow.
It wasn't about standard compliance -- to have more paper work, than coding and having ever growing hardware actually running that kernel.
It wasn't about have kernel today enterprise tomorrow -- money and attention of big (any?) companies.
It wasn't cathedral after all (you know what to read about that).
To have any kind of libc as accessible userspace was a need, GNU project failed to provide. After all FSF lost to Open Source. You must, as i've wrote in my first mail here, build monument to Linus Torvalds. Because of that tiny-crucial part of the operating system -- WORKING kernel.
Yes, now it's easy to talk about POSIX, 64 bits and others things in glibc back in that time. But this have no connection to Linux kernel at all. Do you know why?
Because Cygnus have their business, learn what exact business. And as a very wild guess i can speculate, that glibc by Cygnus was developed using Solaris as f*cking *working* kernel. Feel free to check technical writings on http://people.redhat.com/drepper/.
If Linux by Linus Torvalds and friends after all reached maturity level, when Solaris can be abandoned, again make something useful -- build monument to him!
I'm back coding. -- -o--=O`C info emacs : FAQ #oo'L O info make : not found <___=E M man gcc : not found
Oleg Verych olecom@flower.upol.cz writes:
Let me express it (a wild guess) and save your time for more useful things.
All very interesting, but not very useable for adding anything to the short history I wrote.
I hope you'll understand that I don't have time to go through Drepper's homepage to make deductions from his writings.
If you happen to find a more directly on-topic link, please keep me in mind.
At 19 Jun 2007 14:24:13 +0100, Ciaran O'Riordan wrote:
Having read about the libc5 fork of glibc in Rick Moen's "Fear of Forking" essay, I went looking for more info about this fork since it seems to have healed so well that no one ever talks about it anymore. It also seems strangely underdocumented for a 10 year old event in a community that values putting everything about everything online.
http://fsfe.org/en/fellows/ciaran/ciaran_s_free_software_notes/history_of_gl...
The section heading "glib 2.0" should be "glibc 2.0".
If anyone knows any other sources of info that I didn't reference, I'd be happy to hear about them.
Ask the main players? Roland McGrath, roland@frob.com, the original author of glibc and still one of the maintainers can probably tell you everything you want to know. And you can also ask Ulrich Drepper. There is also this glibc announcement from him, http://sources.redhat.com/ml/libc-announce/2001/msg00000.html (at the end of the announcement):
"The glibc situation is even more frightening if one realizes the story behind it. When I started porting glibc 1.09 to Linux (which eventually became glibc 2.0) Stallman threatened me and tried to force me to contribute rather to the work on the Hurd. Work on Linux would be counter-productive to the Free Software course. Then came, what would be called embrace-and-extend if performed by the Evil of the North-West, and his claim for everything which lead to Linux's success."
Although it's before my time, my guess is that the main reasons of the fork were GNU's cathedral-like development style at that time (see also the egcs fork) combined with the focus on the Hurd instead of Linux.
Jeroen Dekkers
On Tue, 19 Jun 2007 16:39, jeroen@vrijschrift.org said:
fork were GNU's cathedral-like development style at that time (see also the egcs fork) combined with the focus on the Hurd instead of
FWIW, glibc implements large parts of the Hurd and thus one should not wonder that it is quite complex.
Shalom-Salam,
Werner
fork were GNU's cathedral-like development style at that time (see also the egcs fork) combined with the focus on the Hurd instead of
FWIW, glibc implements large parts of the Hurd and thus one should not wondper that it is quite complex.
The Hurd parts are quite simple, and few, if one compares to just the ELF loader.
Jeroen Dekkers jeroen@vrijschrift.org writes:
The section heading "glib 2.0" should be "glibc 2.0".
Fixed, thanks.
Ask the main players? Roland McGrath
Good idea.
And you can also ask Ulrich Drepper.
...
"... Stallman ... embrace-and-extend if performed by the Evil of the North-West..."
That doesn't seem so relevent. It was four and a half years after glibc-2.0 was released, and nothing ever came of Drepper's accusations other than a slashdot story.
Having read about the libc5 fork of glibc in Rick Moen's "Fear of Forking" essay, I went looking for more info about this fork since it seems to have healed so well that no one ever talks about it anymore. It also seems strangely underdocumented for a 10 year old event in a community that values putting everything about everything online.
http://fsfe.org/en/fellows/ciaran/ciaran_s_free_software_notes/history_of_gl...
If anyone knows any other sources of info that I didn't reference, I'd be happy to hear about them.
You could ask the horse directly instead of relying on third-hand information. Roland McGrath (person who wrote glibc 1.x and 2.x) and H.J. Lu who maintained Linux libc4 and libc5.