Slashdot Mirror
Watch Out Linux, GNU Hurd Coming
sfcrazy writes "Debian now has concrete plans to bring GNU Hurd to the larger community. GNU Hurd is expected to be released with the release of Debian 7.0 Wheezy towards the end of 2012 or beginning of 2013. Debian maintainer Samuel Thibault has already produced a Debian GNU/Hurd CD Set with a graphic installer which is available to download."
Duke Nukem Forever actually gets released and now Hurd? Pinch me I must be dreaming!
Hold up, wait a minute, let me put some pimpin in it
Oh lawd! Somebody catch me. I've caught the vapors!
Much like it's long-awaited vaporware cousin, Duke Nukem Forever, the wait will not be worth it.
I don't believe in time. It's a grand conspiracy designed to sell watches.
Will be the year of the Hurd Desktop. 'Nuff said.
One of the very few people to put me on her Slashdot enemies list did so because I made a derogatory statement about the length of the HURD development process. In, as I recall, the year 2000 or 2001. It was a running joke at least five years before that.
Way to be timely and relevant, GNU.
-- I wanna decide who lives and who dies - Crow T. Robot, MST3K
But as for actually running it day-to-day, the Hurd never was relevant simply because it never had the broad driver support that users need. That won't change unless and until Hurd attracts a substantial developer base - but this is a good step in the direction of that.
I am trolling
BSD userland on top of GNU Hurd.
"What the hell do you call an OS like that?"
"I'll call it 'The Aristocrats'"
--
BMO
I swear I saw this headline 20 years ago.
---- GENERATION 26: The first time you see this, copy it into your sig on any forum and add 1 to the generation.
Debian maintainer Samuel Thibault has already produced a Debian GNU/Hurd CD Set with a graphic installer which is available to download.
It speaks volumes that the highlight is the inclusion of a graphical installer. Likely no mouse support though....
Because I think an angry Stallman is a funny Stallman. So I Iook forward to this new Linux variant and I thank Linus Torvalds from the bottom of my heart. There would be no free software movement without you and you cannot be venerated enough. I wonder how Mr. Torvalds came up with Hurd? Oh wait, it is because he is a genius.
Democracy Now! - your daily, uncensored, corporate-free
Who cares? I mean really... we have all the bases covered by Linux and BSD...
If you need a GPLv3 licensed OS for some reason, this will be one. Linux will probably never contain the patent guards Hurd will. That might be important for some folks.
My God, it's Full of Source!
OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
Seriously, it's been, what, 15+ years? The project seems to have two main phases of development:
1) Several months of intense bickering followed by factions breaking off to make half-assed attempts at porting to a more modern microkernel.
2) A year or two of complete silence as the ports are abandoned and a couple of diehards continue to work on Mach.
Where are they now? They've got a couple of novelty builds that almost work reliably enough to play with for a weekend. Big fuckken deal.
They have probably noticed the end of the world is December 21, 2012 and they hope it's true. That way no one will actually notice Hurd was postponed again ...
Nope. All I saw was an organization that maintains a distribution that is the basis for one of the fastest growing Linux distributions offer a distribution based on a kernel made by an organization that maintains one of the most used software compiler suites and userland tools. Please tell me where I should be looking. :P
These comments are my own and do not necessarily reflect the views or opinions of my employer or colleagues...
it is designed to be easier to port than monolithic kernels like Linux.
but Hurd is still Mach and is therefor pretty much garbage (Hurd/L4 died). Hurd lacks the speed of Linux and FreeBSD or the clustering transparency of Plan9.
You'd probably get more out of Plan9 for doing web development than Hurd. Even though Plan9 is probably not a viable platform without significant commercial investment.
“Common sense is not so common.” — Voltaire
FROM: martin-boundary
TO: jlechem
This is a pinch-a-gram. To activate, please pinch yourself.
It's a Mach microkernel with a bunch of daemons and glibc to emulate the UNIX interface.
Way to be timely and relevant, GNU.
What could be greater show of health and vigor than naming your first release "Debian Wheezy?"
"A" decade?
Here's what Linus Torvalds had to say about the GNU OS in 1992
"If the GNU kernel had been ready last spring, I'd not have bothered to even start my project: the fact is that it wasn't and still isn't. Linux wins heavily on points of being available now."
This reminds me of the scene in Austin Powers where the guy screams as a steamroller comes barreling at him at a couple feet per minute. And he stands there and screams for the entire two minutes it takes for the steamroller to reach him and run him over.
Hey! In two years, I'll release the bestest mesh network system ever! I promise!
GNU Hurd is expected to be released with the release of Debian 7.0 Wheezy towards the end of 2012 or beginning of 2013.
A couple of years now. Just like cheap solar panels and sustainable fusion and the replacement for the space shuttle. Just a couple more years now.
How about you call us when it's working?
Seriously, stop telling us what you are going to do. Instead tell us what you have done. One is impressive and the other is not.
Weaselmancer
rediculous.
If Valve really didn't know how to count to 3, we'd have Half-Life 2, Half-Life 2' Champion Edition, Half-Life 2 Turbo, Super Half-Life 2, Super Half-Life 2 Turbo, Super Block-Life 2 Turbo (a puzzle game), the midquel Alpha series (this is either Portal or something connected to it), and finally Half-Life 3.
The question I'd posit at this point is why? Why support a project that was already well on its way to being defunct fifteen years ago? Why support a project whose punchline was "Duke Nukem Forever"? Yes, it would have been an awesome thing in the early and mid-90s, but Linux came along and the GNU userland tools were compiled on top of it, and the rest, as they say, is history. So why on Earth would I want to support Hurd, a project that even if it got a big whack of cash right now probably wouldn't have a fully-functional product in two or three years, and even if it did, it's microkernel architecture would make it slower than modern kernels.
The world's burning. Moped Jesus spotted on I50. Details at 11.
It has always been interesting for me to reflect on the HURD. The issue of mind share was crucial when comparing Linux and the HURD. Back in 1991 (or early 1992, maybe... can't remember that far back) I tried to contribute to the HURD. I had done some work on Mach in an OS course at school and was interested in playing more with it. But since I was pretty much a new grad and didn't have a proven track record working on OS kernels, the HURD team told me to take a hike. Well, they were polite about it, but were clear that they didn't want help from a nobody.
Linux was completely different. Linus may have blasted your code, but he accepted any and all help. This created mind share. For those who weren't around at the time, the whole idea of accepting work from any random joe off the street was a relatively new concept. The "Cathedral and the Bazaar" hadn't been coined by ESR yet and the normal way to do things in the free software world was to have one or two uber programmers hacking away, never seeing the light of day. Now everybody realises that a key indicator of success on a free software project is having an open and unobstructed development process.
The HURD has some good ideas, but their initial attitude killed them. Even though the team is very different now (from what I've hear, anyway -- lost interest in it more than a decade ago), there is really no chance of making a comeback, I think. Enticing eyeballs away from other projects will just be too difficult. Linus' biggest contribution to free software was *not* the Linux kernel, IMHO, but rather the development process that Linux used. He showed everyone how it should be done.
From the minutes of the March 2011 FTPMaster meeting if it's not ready for some sort of release it will be evicted from the main archives.
The TODO list is getting better ... but we shall see.
If that can actually be done safely and efficiently, it also means a non-free driver can't crash the kernel or fuck up other drivers. I would guess there are security implications as well.
Right now, a bug in the nVidia kernel driver on Linux could compromise the security of the entire machine, or crash the entire OS, or flip some bit in some other unconnected kernel system (or userland process), and it's hard enough to debug these things when you do have source code. So wanting an untainted kernel makes a lot of sense.
Don't thank God, thank a doctor!
Arch Hurd is already available so no need to wait for Debian to release anything if you want to try it out. http://www.archhurd.org/
This posting illustrates something very interesting: Why slashdot is irrelevant.
Any community that becomes so ingrained in the belief that it is superior is bound for failure. Because once you start believing no one can be better than you, you start to become complacent. The architecture on which HURD is based is technically superior to Linux. Whether this technical superiority translates to superiority in the marketplace is another issue entirely.
In my opinion the slashdot community consists of a lot of wannabes and not a whole lot of doers. Instead of criticizing and making fun of projects which are new or different why don't you embrace them and welcome them? This is one of the reasons I think the open source community has stagnated in recent years.
GC
Gregory Casamento
## Chief Maintainer for GNUstep
The parent poster illustrates something very interesting: Why having a sense of humor and self-irony is still relevant.
Also, ignoring the "I'm better than you"-irony in your post, "technically superior to Linux" can easily be discussed. I think the kernel community would gladly point out that "the best technology" quickly becomes irrelevant if it's impossible to work with (say if takes two decades of flip-flopping just to release something people can use). And having "developer friendliness" as part of the "rate this technology"-bar is not far fetched, unless you actually want said technology to be: Irrelevant.
Someday maybe a HURD distribution will be released, and someday maybe HURD will surpass the market share of Linux. Until then, chill out and enjoy the source.
The architecture on which HURD is based is technically superior to Linux.
Citation needed.
This posting illustrates something very interesting: Why slashdot is irrelevant.
Any community that becomes so ingrained in the belief that it is superior is bound for failure. Because once you start believing no one can be better than you, you start to become complacent. The architecture on which HURD is based is technically superior to Linux. Whether this technical superiority translates to superiority in the marketplace is another issue entirely.
In my opinion the slashdot community consists of a lot of wannabes and not a whole lot of doers. Instead of criticizing and making fun of projects which are new or different why don't you embrace them and welcome them? This is one of the reasons I think the open source community has stagnated in recent years.
GC
Because taking the piss is far more amusing to us in our juvenile little minds.
In all seriousness though, the big problem is that GNU Hurd has just been going on too long. You might notice that many people are comparing it to Duke Nukem, this is because they have both been successively over hyped for too many years. It is like people crying wolf, eventually the would be rescuers just stop listening and let you get eaten.
I started reading this thinking that GNU Hurd had finally found some developers an was on course for a stable release in the near future. After looking around the site it seems that you only have 4 or 5 active developers and are in dire need of more people to make the Wheezy release. If this is the case then try and ask the community for help, cap in hand with humility. You are far more likely to bring developers to the system by that than by simply posting a projected release date which may or may not be achievable.
You are right though when you say the slashdot community has changed a great deal as it certainly has. But some of the people here are still exactly the people you would like to bring to your projects, either GNU Step or Hurd or whatever. The trick is to appeal to them and ignore the mass of immature wanna bees you are so critical of.
The whole problem with hurd has never been a technical shortcoming, it has always been that the people leading the project lacked the people skills needed. Thats certainly not to say that Linus is perfect in this regard, but something certainly made more people throw time at his pet Linux project all those years ago.
I dont read
Are you trying to illustrate your words by acting as if you were yourself really "superior" ?
I heard about hurd long time ago and it was already a long time project. I heard it will soon be released so many time that I can't count them. I even actually spoke with people working on it (about ten years ago) that were assuring me that the project was on the run for a stable release.
Ten years later, I'm acting as supervisor for student writing their own kernels every year: in 4 years of activities I have seen about 7 kernel projects reaching an "interesting state", and those kernel were all "experimental" in their own way: micro-kernel, coded in some specific language (D, OCaml ... ), fully modularized, coded for exotic architectures ... All are single-man project done by students.
Booting and reaching the state were drivers and userland are the next checkpoint is not so hard, even when you deal with "new inner architecture". But keeping a project really active so that you reach a stable state, is much harder, and it seems that hurd fails on that matter.
Hurd might have been "new" twenty years ago, but for now, it is just another not-working micro-kernel.
(oh, mind your respect, should I talk about GNUstep ?)
Marwan Burelle co-Head of EPITA's System Laboratory
I am aware of the joke. However, the joke doesn't provide memorable version nicknames.
Why is it superior? Just because it's a microkernel?
Microkernels were the darling of OS research for almost the entirely of the 90's. But by the end of the 90's, most researchers had had enough. The alleged gains in configurability, reliability, security, and so on never materialized; but what never disappeared was the fact that they were stinking slow. Context switching is a fundamental limitation of such an architecture. And from what I've heard, a lot more complicated to program -- which leads to more programming errors and ugly performance hacks to compensate for any potential increase in reliability, security and so on they might have gained.
It's possible that Hurd has managed to overcome these limitations. But it has definitely earned its reputation of being slow and cumbersome; if that has changed, the burden of proof is on the Hurd community.
There are a few True Believers out there, still working on Hurd and Minix and L4 and the like, but they have yet to produce anything shown to be worth using.
I think the fact that Andrew Tanenbaum riduculed Linux in 1993 for being an "outdated architecture", when Minix just got paging working last year after 20 years of development, encapsulates my point completely.
TCP: Why the Internet is full of SYN.
Now that you put it that way....
Excuse me for not getting your joke.
I'm going to need surgery to remove my hand from my forehead now.
--
BMO
Followed the link until I here:
/proc/version
;)
http://www.gnu.org/software/hurd/hurd/running/debian.html
So to try it I downloaded and ran in in qemu as per the instructions.
Playing around a bit I noticed this:
root@debian:~# uname -a
GNU debian 0.3 GNU-Mach 1.3.99/Hurd-0.3 i686-AT386 GNU
root@debian:~# cat
Linux version 2.6.1 (GNU 0.3 GNU-Mach 1.3.99/Hurd-0.3 i686-AT386)
So maybe you're right in calling it Linux
This posting illustrates something very interesting: Why slashdot is irrelevant.
Any community that becomes so ingrained in the belief that it is superior is bound for failure. Because once you start believing no one can be better than you, you start to become complacent. The architecture on which HURD is based is technically superior to Linux. Whether this technical superiority translates to superiority in the marketplace is another issue entirely.
In my opinion the slashdot community consists of a lot of wannabes and not a whole lot of doers. Instead of criticizing and making fun of projects which are new or different why don't you embrace them and welcome them? This is one of the reasons I think the open source community has stagnated in recent years.
GC
I agree.
"Ignorance more frequently begets confidence than does knowledge"
- Charles Darwin
For those too lazy to use a search engine:
http://www.gnu.org/software/hurd/community/weblogs/ArneBab/technical-advantages-of-the-hurd.html
I have never seen such ignorant arguments:
- Conflation of development time with product quality: "Minix just got paging working last year" Last I heard, quality products take MORE time to develop, not less.
- Complaints of "inefficiency" when the target platform has 10X the necessary compute power for the task at hand.
- Complaints about "long development time" when compared to the 20+ years that it has took for BSD to achieve commercial success in the market as OSX.
If any of you people would actually stop to read the hurd design docs you would realize that it has already had influence on your desktop. FUSE and SELinux are bolted-on implementations of concepts that were first fleshed out and implemented in the hurd.
I made an assumption because GGP didn't explain himself. If he doesn't want people to misunderstand him then he needs to be more explicit.
In any case, I didn't tear the idea apart. I said that historically, the experience of people using microkernels has been (1) they're really slow, and (2) they're more complicated to program because of the isolation / message passing architectures, and thus more prone to bugs. I admitted that I didn't know anything about Hurd, but said that the burden of proof is on the Hurd community to show that these problems are no longer valid. That was an open invitation to Hurd supporters to bring out the facts and set me straight.
What I've gotten so far (including you) is mostly people saying that CPUs are so fast these days that the extra overhead doesn't matter. No explanation of how Hurd is technically better (other than "it's a microkernel"), no comparison of difficultly in design or programming in the full microkernel environment compared to a monolithic kernel like Linux or the *BSDs, and no real quantification of ways in which the extra overhead / complexity is worth the cost.
No, I haven't read the design docs. It's not my job to look at every half-baked idea out there to see if it's worth considering. And I've heard a lot of microkernel fanboi stuff in my time; I'm skeptical and a bit jaded wrt microkernels. If you and/or the Hurd community want Hurd to get more "buy-in" from the FLOSS community in general, you need to be better evangelists -- understand where the skeptics are coming from, and come armed with facts and arguments that are persuasive and attractive. "The architecture on which HURD is based is technically superior to Linux" is not an attractive or informative statement.
TCP: Why the Internet is full of SYN.
Developers of HURD think HURD is superior to the competition. Film at 11.
I guess we should have specified that we wanted an INDEPENDENT look at whether the HURD was superior.
>Hurd was a Victim of Good Enough
Moreso a victim of the Perfect being the Enemy of the Good (assuming you're willing to give it that much credit).
And it has given new meaning to "next generation" software, coming a full *human* generation later . . .
hawk
...they need to go write their own OS where proprietary drivers are okay, because we Linux users don't want them.
Fuck you, you do not speak for all Linux users.
I would much prefer an open source driver to a proprietary one, all things being equal. All things are not equal. As cool as AMD has been lately, their proprietary Linux drivers still have far better 3D performance than the open ones.
More importantly, I run proprietary software on Linux, even proprietary software I've paid for! I'm ok with that.
I would still rather my system be open source, and I would especially like it if the proprietary stuff I run (even drivers) was properly sandboxed. I find myself wondering why I should be forced to trust this gigantic binary blob from nVidia more than I trust the JavaScript running on any random website. I understand the technical reasons why this is needed right now, but if something like HURD can solve them, I'm all for that.
Don't thank God, thank a doctor!
I did some research, and according to this article, although OSX does use Mach, it is nonetheless not a microkernel:
TCP: Why the Internet is full of SYN.