Courtesy of LHB at July 14, 2011 7:56 PM.
It’s worth reading the follow-up comments on LHB (you may have to skip the odd nigger-related or (anti-)Marxist diatribe), because they nail this one.
I’m trying to remember what the Hurd was supposed to be in the first place (apart from the ultimate RMS wet-dream), but I seem to remember that they co-opted the Mach kernel fairly early on and then failed, miserably. The Mach kernel, you see … and I have a lot of time for the Mach kernel and subsequent work in that direction … does not run on toasters. It also requires significant architecture and design capability.
So, not ideal for FOSS, then.
No matter: the Loons have brought it back from the dead in, er, Debian 7.0 Wheezy. Stuff Logo contests; they can’t even manage Naming contests these days.
The fun these people are going to have with a total disregard of the concept of “user space” drivers. Or ABI/API stability. Or, well, practically anything that Mach is designed to do and Linux isn’t.
Which of course brings us to another question. Exactly how much of the Linux Desktop Environment experience would survive this?
Toasters! You know they make sense.
Also traffic lights and routers. And, of course, Supercomputers.
Comments
Duke Nukem Forever has finally arrived and now HURD? The world really will end in 2012.
Not if the various Linux calendar applications have anything to do with it.
Actually, that’s an interesting thought.
If the Loons have coded it right, perhaps there is a multiverse, after all.
They’re back to the Mach kernel again now?
Well, the original idea was to base the Hurd on the Mach kernel, and they kept trying to shoehorn the AwesomeHurdAbilities™[1] into it for about two decades. Eventually, They decided that Mach was too old (I mean, who would use a kernel that was last released a decade ago?). They then decided to use the L4 kernel, but also failed to release anything meaningful. Then, someone had the amazing idea of using the Coyotos kernel because it had “capabilities” or something. Guess what? They cancelled that before even writing a single line of code.
The last plan was creating their own microkernel from scratch (called Viengoos). You heard that right, those great coders who can’t even get an existing mature kernel to work were going to create one from scratch.
Yeah, I’m sure they’ll get it to work by “the end of 2012”.
——
“perhaps there is a multiverse, after all.”
Yes, and it’s full of evil non-free programs![2]
[1] http://www.gnu.org/software/hurd/advantages.html
[2] https://help.ubuntu.com/community/Repositories/Ubuntu
How can you NOT get L4 to work? What is it about HURD that makes it so they can’t use L4 for it?. After you have a pager and a root task working everything becomes second nature (and those are already provided in L4Re, including a working c++ compiler).
I registered on slashdot only to upvote this:
“Much like it’s long-awaited vaporware cousin, Duke Nukem Forever, the wait will not be worth it.”
Anyway, I never got the point behind Hurd AND Linux. Didn’t BSD-unencumbered and later FreeBSD provide a perfectly working kernel? And even if the FSF loons were dissapointed that they were not released under the holy GPL license, they could have always forked them, add a couple of frills, and release them as their own.
Now, how about building a graphics and audio stack that doesn’t suck?
BTW, does anybody know how you can “upvote” someone on Slashdot?
I just played with the “modifiers”, whatever a “modifier” is. Linux usability has finally found it’s way to websites!
HURD, or more practically, TURD has nothing to offer as far as I can tell. Poking around I found a LiveCD of HURD. In the middle of booting up, it threw a panic fit and rebooted. None of the various boot options worked… every time a panic. If I want crap like that, I’ll stick to Linux.
“How can you NOT get L4 to work?”
By having your head stuck up your bottom.
Well, this is the official reason1, which just proves my point: “partly due to [the Hurd developers] not cosidering L4 suitable for implementing a general-purpose operating system on top of it”.
At least they have the gut to admit that it was also “because of deficiencies in the original Hurd’s design”.
Even if Debian managed to patch up a somewhat-working Hurd in 2012-2013, no one will consider it a “release” until the FSF starts running it on its servers, and recommends it instead of gNewSense/Trisquel/etc.
[1] http://www.gnu.org/software/hurd/history/port_to_another_microkernel.html
@DigitalAtheist
Did you boot it in QEMU? I think QEMU is the only supported platform at the moment.
@Kommenter:
“How can you NOT get L4 to work?”
It takes a million eyes, and a soupcon of obdurate stupidity, but I guess it’s worth it in the end.
A bigger question is why anybody would take the Hurd over, ooh, dunno, say L4.
GNU/Linux is fine for now, but GNU/Hurd has the potential to be a much better OS.
You must be signed in to leave comments.