Announcement

Collapse
No announcement yet.

GNU Hurd 0.5, GNU Mach 1.4 Released

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • GNU Hurd 0.5, GNU Mach 1.4 Released

    Phoronix: GNU Hurd 0.5, GNU Mach 1.4 Released

    In celebration of the GNU project's 30th birthday, GNU Hurd 0.5 has been released along with GNU MIG 1.4 and GNU Mach 1.4...

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    fantastic

    fantastic project after 30 years continue not done

    Comment


    • #3
      Unbelievable! Where do this guys get the motivation to continue to develop this non-sense?
      GNU still lives in the 80s, with all its GUILE and APL crap which nobody will ever use.

      Comment


      • #4
        Originally posted by Andrecorreia View Post
        fantastic project after 30 years continue not done
        GNU itself is celebrating its 30th birthday. Hurd has been around for 23 years. At what point do you consider it done?

        Comment


        • #5
          Originally posted by DanL View Post
          GNU itself is celebrating its 30th birthday. Hurd has been around for 23 years. At what point do you consider it done?
          When it will usable like free bsd for example....

          Comment


          • #6
            Originally posted by pandev92 View Post
            When it will usable like free bsd for example....

            Did you try Debian GNU/Hurd ?

            Comment


            • #7
              Originally posted by BO$$ View Post
              It's dead people. Bury it and be done with it. It's a failure. Accept it. Move on.
              the biggest problem I think is driver support, but maybe I am wrong here.
              But there are some features, that are not there in linux, really nice features. So the question is is that a design thing or why is linux not doing that.

              So you could say after 30 years linux cant do stuff that hurd can do, so its a failure
              Of course thats a bit of a joke, but bsd has the wrong lisence and is basicly completly patented as far as I know? So its no alternative.

              I mean the advanteges I refer to to make really true that "everything is a file" even if its a ftp server or something. Yes with mounting u can do something closely, but thats not the same.
              The only programm that could do something similar yet (a link to a ftp file as example) is also another old gnu project emacs with org-mode can do something like that. a Text-link to a ssh file even with su rights or somethign like that.

              So it seems linux is able to do that, but only with old gnu-power ^^ but of course emacs is dead also right? ^^

              Comment


              • #8
                Originally posted by blackiwid View Post
                the biggest problem I think is driver support, but maybe I am wrong here.
                But there are some features, that are not there in linux, really nice features. So the question is is that a design thing or why is linux not doing that.

                So you could say after 30 years linux cant do stuff that hurd can do, so its a failure
                Of course thats a bit of a joke, but bsd has the wrong lisence and is basicly completly patented as far as I know? So its no alternative.

                I mean the advanteges I refer to to make really true that "everything is a file" even if its a ftp server or something. Yes with mounting u can do something closely, but thats not the same.
                The only programm that could do something similar yet (a link to a ftp file as example) is also another old gnu project emacs with org-mode can do something like that. a Text-link to a ssh file even with su rights or somethign like that.

                So it seems linux is able to do that, but only with old gnu-power ^^ but of course emacs is dead also right? ^^
                I don't really see the point, here. Linux is GPL, just as Hurd. If the only point is to bring the "everything is a file" idea closer to reality, then do it in Linux, instead of wasting efforts duplicating the work of everything else. I believe such a feature could be achieved as a build time switch, not even disturbing everyone else's way. Even if they don't want to work with Linux people because of whatever reason, they could as well fork it, because, again, the license is GPLv2 or above, used in every GNU project.

                Comment


                • #9
                  Originally posted by mrugiero View Post
                  I don't really see the point, here. Linux is GPL, just as Hurd. If the only point is to bring the "everything is a file" idea closer to reality, then do it in Linux, instead of wasting efforts duplicating the work of everything else. I believe such a feature could be achieved as a build time switch, not even disturbing everyone else's way. Even if they don't want to work with Linux people because of whatever reason, they could as well fork it, because, again, the license is GPLv2 or above, used in every GNU project.
                  But this modularity is basicly the point of a micro-kernel. So think of you have 1000 different types of things, in a micro kernel as far as I understand it, you have for every mini-thing a seperate "driver" or prozess.

                  Like I said it should also be possible, but I think its more difficult to it there, a monolyth kernel is not so good in such stuff.

                  There is ftpfs as example in hurd you have a server for everything this servers talk to each other and you can pipe most stuff from one to another.


                  Like I said I am no expert on that, and I do not develop any os ^^. I think in the long run it could be the better aproach I know it sounds strange, but look what else did get replaces, now its the x server from one evolutional replace another x11 X11R6 -> Xorg, now its wayland. That took also 30 years or so? 40?

                  Why should that not happen one time with linux. a microkernel is more modular, and more modularity is better, its sometimes harder to get there, but if you optimised finaly everything else where will you optimise then.

                  But yes it can take another 30 years maybe till a new system will be the main operation system. But bsds userbase is also 1/100 of the linux user base but still its around and kicking. So why not having such small thing for geeks around that in the long run will be mainstream maybe.


                  I am also a bit pissed about gpl2 shit, what we have with android nsa-spyware hurts me, its in reallity the biggest crap availible, even windows on tablets/phones is faster/more efficient, and also the apple stuff.

                  And btw why have arm systems no bios or efi or even better coreboot on them. I hate the flashing process. its computers why are they not so open like other computers, but ok thats OT now.

                  Comment


                  • #10
                    Originally posted by blackiwid View Post
                    But this modularity is basicly the point of a micro-kernel. So think of you have 1000 different types of things, in a micro kernel as far as I understand it, you have for every mini-thing a seperate "driver" or prozess.
                    Yes, that's the idea of the micro kernel. But again, they could match their own Mach kernel with a FreeBSD forked (the fork would be to use it under the GPL, since that's one of the hard requirements of the project) monolithic server to have features, so they could have an usable OS from the beginning and then slowly modularize into smaller servers. IIRC, Darwin (OS X open source infrastructure, including the kernel) uses such kind of hybrid. I think they used Mach and then switched to something called L4 as microkernel (I didn't look up for what L4 actually is, so it might be just the same thing but as a version or something like that), and FreeBSD modified kernel as a server for the microkernel.

                    Like I said it should also be possible, but I think its more difficult to it there, a monolyth kernel is not so good in such stuff.
                    I don't know why you say this. I don't actually see how those features relate to each other. A microkernel is probably worse when you try to convince people to follow a specific rule, like exposing everything as a file, since they can just write servers without even looking at such rules, as long as they find the API they need to follow. Or did you mean the flexibility on switching behaviors?

                    There is ftpfs as example in hurd you have a server for everything this servers talk to each other and you can pipe most stuff from one to another.
                    I don't understand this part.

                    Why should that not happen one time with linux. a microkernel is more modular, and more modularity is better, its sometimes harder to get there, but if you optimised finaly everything else where will you optimise then.
                    Optimization and microkernels aren't really hand in hand, you know? A microkernel, by its own architecture, implies a lot of extra calls and inter-process communication. It's its nature. It's focus is simply not speed, but reliability and security. Your system will probably be a lot more stable using a microkernel, but an optimized microkernel + optimized servers will always be slower than an optimized monolithic kernel providing the exact same features (and you could limit to that exact same features by compiling it with only those you need). For uptime critical applications, go with the micro kernel. For time critical, a monolithic kernel might be wiser.

                    But yes it can take another 30 years maybe till a new system will be the main operation system. But bsds userbase is also 1/100 of the linux user base but still its around and kicking. So why not having such small thing for geeks around that in the long run will be mainstream maybe.
                    The small thing for geeks, yeah, I like to try new things, too. I just don't see a real niche, since AFAIK the main reason for Hurd is not having a microkernel, but allowing to have a fully free OS, and Linux + GNU userland already provides such an OS. As a toy OS, there are a lot, and another one will not hurt anybody

                    I am also a bit pissed about gpl2 shit, what we have with android nsa-spyware hurts me, its in reallity the biggest crap availible, even windows on tablets/phones is faster/more efficient, and also the apple stuff.
                    What's the problem with GPL2? Android's specific thingies aren't GPL, IIRC, but MIT. You can't close up GPL2 software, since it's copyleft. Performance and licensing aren't correlated, or if they are, I haven't seen any proof.

                    And btw why have arm systems no bios or efi or even better coreboot on them. I hate the flashing process. its computers why are they not so open like other computers, but ok thats OT now.
                    Yes, that's really off topic, as nobody was discussing anything related, and I have no idea why they are so closed and all of that. The flashing process, AFAIK, has nothing to do with them having a BIOS, but with them storing the OS in an EEPROM. This could be done even if using a fully open stack.

                    Comment

                    Working...
                    X