Announcement

Collapse
No announcement yet.

Ubuntu 9.10 Off To A Great Performance Start

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

  • #61
    Originally posted by deanjo View Post
    Perhaps we should apply the same philosophy to hardware in the kernel too then. Think of all the additional talent that could be used in focusing on current hardware instead of maintaining old legacy hardware. Think of how much easier it would be to maintain the kernel if all that old hardware support was scrapped.
    I really fail to see how your logic works. You're taking an argument that relates to closed source software _leaching_ on the kernel and transferring it to open source software _within_ the kernel.

    The main point raised here in contra-stable-API camp was that catering to closed source software isn't sustainable and in fact not even the kernel's job. It is, however, one of the kernel's declared jobs to support even very dated software.

    Also, as for old hardware, why shouldn't one use the 2.2.* / 2.4.* / 2.6.16.* branches? Old hardware can anyway not use the latest features, therefore it doesn't _really_ make sense to apply the newest kernel to it. As long as you port back the security patches, you're well off.

    On a more general note, I would say, as long as something is open source it is per definitionem (more) maintainable.

    Comment


    • #62
      Originally posted by susikala View Post
      I really fail to see how your logic works. You're taking an argument that relates to closed source software _leaching_ on the kernel and transferring it to open source software _within_ the kernel.

      The main point raised here in contra-stable-API camp was that catering to closed source software isn't sustainable and in fact not even the kernel's job. It is, however, one of the kernel's declared jobs to support even very dated software.

      Also, as for old hardware, why shouldn't one use the 2.2.* / 2.4.* / 2.6.16.* branches? Old hardware can anyway not use the latest features, therefore it doesn't _really_ make sense to apply the newest kernel to it. As long as you port back the security patches, you're well off.

      On a more general note, I would say, as long as something is open source it is per definitionem (more) maintainable.
      Maintaining an API to sustain is extremely doable without breaking previously working software. 90+ percent of the world runs on such a system and it does fine.

      As to why to not use an old kernel that's really simple. Old kernels are maintained for security but outstanding bugs are left as "Won't Fix" because the fixes are found in the current version of the kernel. Plus again in that situation older kernels do not have support for newer hardware that maybe required to be replaced. Trying to run a 2.6.16 kernel for example on most modern motherboards simply will not work at all.

      opensource is only more maintainable if there is someone willing and qualified to do so. Same can be said about closed source. In either case if there is nobody interested in doing so with the qualifications your hooped either way. There are millions of unmaintained open and closed projects out there sitting around because nobody has the interest or skills to do so. Your basing your conclusion on ideal circumstances, then again under ideal circumstances a closed source provider would support their solution as well indefinitely as well.
      Last edited by deanjo; 22 May 2009, 10:10 AM.

      Comment


      • #63
        Maintaining an API to sustain is extremely doable without breaking previously working software. 90+ percent of the world runs on such a system and it does fine.
        Agreeing to that statement is equal to declaring that the large bulk of Linux developers are driven by their ego. While I wouldn't say this is necessarily not the case, I can't agree with that statement, since Linux is in too many respects simply superior to every other operation system out there (just look at fefe's benchmarks, it says it all). So, "extremely doable" or not, it works better than stable APIs.

        The next sentence is even more ludicrous. The majority of this world's servers run Linux _and_ work fine. The majority of this world's personal computers run Windows and are slow, infested with viruses and as stable as cardboards. If that is what you term 'does fine', then your expectations of software are really low.

        As to why to not use an old kernel that's really simple. Old kernels are maintained for security but outstanding bugs are left as "Won't Fix" because the fixes are found in the current version of the kernel. Plus again in that situation older kernels do not have support for newer hardware that maybe required to be replaced. Trying to run a 2.6.16 kernel for example on most modern motherboards simply will not work at all.
        I don't even need to comment on that. First you say that removing support for old hardware would speed development up, then when I actually _agree_ and say the people who have old hardware should keep to old(er) kernels, you say it won't work since those lack support for new hardware. I'm sorry, but can't you maintain a thread of thought?

        opensource is only more maintainable if there is someone willing and qualified to do so. Same can be said about closed source. In either case if there is nobody interested in doing so with the qualifications your hooped either way. There are millions of unmaintained open and closed projects out there sitting around because nobody has the interest or skills to do so. Your basing your conclusion on ideal circumstances, then again under ideal circumstances a closed source provider would support their solution as well indefinitely as well.
        I largely agree on that. Only the difference is: open source guarantees the code is open and work can picked up at any time. Of course, if the source is badly commented then it'd be a pain in the arse, but it's still better than what, being able to do nothing?

        Comment


        • #64
          Seriously:
          Originally posted by http://www.kroah.com/log/linux/ols_2006_keynote.html
          When a company says that they need to "protect their intellectual property", that's fine, I and no other kernel developer has any objection to that. But by the same token, you need to respect the kernel developers intellectual property rights. We released our code under the GPL, which states in very specific form, exactly what your rights are when using this code. When you link other code into our body of code, you are obligated by the license of the kernel to also release your code under the same license (when you distribute it.)
          [...] Some companies try to skirt the license of the law on how they redistribute their closed source code, forcing the end user of it to do the building and linking, which then causes them to violate the GPL if they want to give that prebuilt module to anyone else. These companies are just plain unethical and wrong.
          Closed source modules are illegal. So there's no point in supporting them with a stable api (but pretty many points against it, which have already been mentioned in this thread).
          End of discussion.

          Comment


          • #65
            Originally posted by susikala View Post
            Agreeing to that statement is equal to declaring that the large bulk of Linux developers are driven by their ego. While I wouldn't say this is necessarily not the case, I can't agree with that statement, since Linux is in too many respects simply superior to every other operation system out there (just look at fefe's benchmarks, it says it all). So, "extremely doable" or not, it works better than stable APIs.
            I'm sorry but a model that continuously has to break existing infrastructure to progress is not superior at all from a end user perspective. This is why we do not see mass adoption in the average user desktop space.

            The next sentence is even more ludicrous. The majority of this world's servers run Linux _and_ work fine. The majority of this world's personal computers run Windows and are slow, infested with viruses and as stable as cardboards. If that is what you term 'does fine', then your expectations of software are really low.
            Who is talking about a static environment that webservers enjoy? I love how fans such as you always have to pull the "server card" when they have nothing else to fall back on.


            I don't even need to comment on that. First you say that removing support for old hardware would speed development up, then when I actually _agree_ and say the people who have old hardware should keep to old(er) kernels, you say it won't work since those lack support for new hardware. I'm sorry, but can't you maintain a thread of thought?
            I wasn't saying any such thing I was applying your reasons to breaking backwards support to the same application in the linux kernel. By removing such support in the kernel you lose one of linux's biggest features which is it's ability to work with hardware from multiple generations. This is also what API's should strive to achieve instead of having to go back to the drawing board every few months never actually completing one solution before progressing to the next.

            I largely agree on that. Only the difference is: open source guarantees the code is open and work can picked up at any time. Of course, if the source is badly commented then it'd be a pain in the arse, but it's still better than what, being able to do nothing?
            For the most part closed source can be purchased and continued to be developed by a entitiy wanting to do so as well. It happens all the time in closed source.

            Comment


            • #66
              Originally posted by deanjo View Post
              I'm sorry but a model that continuously has to break existing infrastructure to progress is not superior at all from a end user perspective. This is why we do not see mass adoption in the average user desktop space.
              See my last post and Apopas' link following it. You obviously have no idea what you are on about.

              Originally posted by deanjo View Post
              For the most part...
              What? How do you know the attitudes of "most" closed source developers? What if they don't want to sell? How is this in any way equivalent to open source, where code is always available to be modified, forked etc.?
              Originally posted by deanjo View Post
              ...closed source can be purchased and continued to be developed by a entitiy wanting to do so as well.It happens all the time in closed source.
              I don't think it does. Got any examples? As the entity doesn't necessarily have the cash to pay whatever the asking price is for the project, I think this more often results in people reinventing the wheel.

              I know your scenario often happens in open source. Except the purchasing bit . This means that even if a single developer is interested in a project they are free to continue development without hindrance. See, for example, mpg123. I have never seen something similar happen with closed source software.

              Comment


              • #67
                Originally posted by krazy View Post
                I know your scenario often happens in open source. Except the purchasing bit . This means that even if a single developer is interested in a project they are free to continue development without hindrance. See, for example, mpg123. I have never seen something similar happen with closed source software.
                Tremendous example is Blender which needed to be bought by the community and become open-source to be able to become advanced and mainstream competitive...
                Anyway, we can argue till eternity for such things, the matter is just one. Linux is superior to any other proprietary filesystem, the benchmarks is out there as well as the technology is to test for yourself if you want, that means that the current model of creation indeed works! Some may argue that we didn't test the other way to be sure that it would work better like that but hey window$ and maco$x do that and "follow" Linux in everything but mass people adoption! But seriously is the way kernel is built the reason for that? I mean if for example proprietary quality games are essential for guys around the world to use Linux then why Loki closed its doors? Come on guys, other things are responsible for Linux's inpopularity, financial mainly and because the absolute majority of people don't feel comfortable with instant changes in their life, that's human nature and for that we'll see Linux slowly but surely to take its place in the world's desktop sales.

                Comment


                • #68
                  Originally posted by krazy View Post
                  See my last post and Apopas' link following it. You obviously have no idea what you are on about.
                  I know exactly what I'm talking about. GKH's rant is more of a religious statement then anything having to do with a technical perspective.

                  What? How do you know the attitudes of "most" closed source developers? What if they don't want to sell? How is this in any way equivalent to open source, where code is always available to be modified, forked etc.?
                  Closed IP is always for sale. Believe me if a company can make a last buck off a dead product they will. Just this year the software company I work for picked up six pieces of unmaintained IP and sold two and this is from a relatively small company.

                  I don't think it does. Got any examples? As the entity doesn't necessarily have the cash to pay whatever the asking price is for the project, I think this more often results in people reinventing the wheel.
                  Oh lets pick a couple shall we? Let's see DOS comes to mind, Wordperfect, in fact just go ahead and type Microsoft squires or Apple acquires and you'll find many high profile scenario's where this happens on a regular basis.

                  I know your scenario often happens in open source. Except the purchasing bit . This means that even if a single developer is interested in a project they are free to continue development without hindrance. See, for example, mpg123. I have never seen something similar happen with closed source software.
                  Without hinderence maybe to code, except for the fact again you are hoping someone with the knowledge and interest is willing to do so. It also can hinder you towards what application of use is depending on the license. This however is not a viable solution for most end users who would end up paying more for the continued development of a solution where the market is small. If it wasn't for the almighty large corporate dollar paying for most of linux's development it would be decades behind any other solution. Opensource likes promote itself on a worse case scenario FUD towards closed source. While that worst case can happen, it is by no means the defacto standard.

                  Comment


                  • #69
                    Originally posted by Apopas View Post
                    I mean if for example proprietary quality games are essential for guys around the world to use Linux then why Loki closed its doors?

                    Oh man that is a can of worms that you don't want to get into. It being essential or not isn't really why Loki closed the doors. Piss poor management and some crooked people are responsible for that mess. I'll let Svartalf rant on about that.

                    Comment


                    • #70
                      Are there actually some sensible corps that don't charge six figures for dead dos-era software?

                      Comment

                      Working...
                      X