Announcement

Collapse
No announcement yet.

Reasons Why You Should Not Use FreeBSD

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

  • #31
    Originally posted by kraftman View Post
    No, the biggest problem are BSD fanboys who have no clue about Linux and who just troll about it. foolBSD people are the best example. They can't live without Linux.
    They have no clue about Linux, but they can't live without it?!
    Most BSD fanboys use or used Linux too. Most blind Linux fanboys never used BSD.
    BSD fanboys know more about Linux, then Linux fanboys know about BSD.
    Many Linux/BSD fanboys can't live without Windows.

    Comment


    • #32
      Originally posted by disi View Post
      I am not sure, if this is an argument. In the company I work for, they always say "but I cannot get professional support for free Linux distributions if something goes wrong". They want me to use VMWare Hypervisor over KVM because of this, where the ESXi Hypervisor lacks driver support for recent hardware worse than FreeBSD
      Was that smile inidcating a joke or that you thought your company is idiotic?
      Either way, just point your it boss to red hat's costumer site. They do business with truly massive companies as well as the US government.
      I really don't understand why kvm hasn't had more up take thus far...

      Comment


      • #33
        Originally posted by kraftman View Post
        UFS, ext[1] and fat are real crap. If zfs is production ready in BSD then it's great, but it is not so good for desktops.
        Then ext4 is based on crap. Why ZFS isn't good for desktops?


        Originally posted by kraftman View Post
        FreeBSD contains 30 year old code, so it's hard to tell if it is younger. When comes to automation I meant things like udev. Atheros is mainly OpenBSD merit. It's hard to find anti-BSD trolls at Linux distribution forums and I didn't ever find a Linux dev who trolls about BSD. Devs from FreeBSD trolls a lot about Linux.
        All code was rewritten many times, also because licensing problems.
        There are anti-BSD trolls at this forum.
        Linux devs troll BSD with constant compatibility braking of shared userland.

        Comment


        • #34
          Originally posted by liam View Post
          Was that smile inidcating a joke or that you thought your company is idiotic?
          Either way, just point your it boss to red hat's costumer site. They do business with truly massive companies as well as the US government.
          I really don't understand why kvm hasn't had more up take thus far...
          It's a work place and not my company.
          What I meant: people are scared if they cannot sue/blame anyone else.
          Take Amazon cloud: Companies upload data to their servers(mainframes) and if it gets lost they are out of business, BUT they can blame/sue Amazon for it.

          Comment


          • #35
            So little helpful posts, so much hate. Best one was the one written by Eisnefaust (btw, is that a typo?), but nobody seems to care.

            Comment


            • #36
              copyleft vs. BSD license

              Originally posted by vertexSymphony View Post
              2) Well, it's your own opinion and taste, we had enough flame wars about licensing ... and I could't see a SANE argument against the license , none (no, repeating other peoples argument without the understanding doesn't count)
              Is that so? Let me present what I consider a sane argument against the so-called "permissive" license then and I'd be interested if you consider it sane too or hear your counter-arguments.

              I do care about the 4 freedoms as defined by FSF and believe everyone should have those as it nurtures long term innovation, empowers users to do what they want, limits abuse of the userbase by the developer and maximizes code reuse among other key advantages:

              • The freedom to run the program, for any purpose (freedom 0).
              • The freedom to study how the program works, and change it so it does your computing as you wish (freedom 1). Access to the source code is a precondition for this.
              • The freedom to redistribute copies so you can help your neighbor (freedom 2).
              • The freedom to distribute copies of your modified versions to others (freedom 3). By doing this you can give the whole community a chance to benefit from your changes. Access to the source code is a precondition for this.


              The so called "permissive" licenses enable anybody to take those freedoms away. So only someone who doesn't believe those freedoms are essential can advocate for it as he doesn't mind them being taken away. It means little that he himself perhaps doesn't take them away directly (albeit he could at any moment change his mind) but "just" enables anyone else to do so - there is no significant difference there.

              So that's an argument from the perspective of software freedom. In the case someone doesn't care for software and user freedom (which is a clear deal-breaker for me already) there is also the point about innovation.

              With GPL the community gets all the innovation stemming from it and building on top of it, there is no situation where some entity can take the code away, add some polish and never contribute back - if you want to build upon the thousands/millions of hours others spent improving the code that you get for free, it's only fair you contribute back any modifications you make. And this model works great for Linux, on the other hand maybe that's why BSD is being left behind, there is no such balance in "permissive" land - anyone who wants to "invest" in it can just take the code and never contribute back - that doesn't work long term either for BSD or the ones forking it into proprietary products (but they keep trying - and failing wasting valuable coding effort that could be used to improve the underlying system for all).

              The best testament to this is indeed the contrast between the adoption of BSD vs. Linux - copyleft works, it works great - if you're looking for long term benefit that is, it doesn't work for short-sighted entities who want to take away the essential freedoms it provides for the shot-term "benefit" of being able to monopolize their contribution.

              So two interrelated arguments:

              1) "permissive" licenses do not protect the 4 freedoms that are essential to benefit the users and long term innovation (the next point)
              2) copyleft provides long term innovation vs. short-term benefits of "permissive" licenses such as BSD which hinder innovation in the long term

              note: why do I keep putting "permissive" in quotation marks? It's because I consider it grossly misleading - the only thing it "permits" you to do is to take away freedom from others - so in effect it "permits" you to restrict others. I do not consider a permission to restrict others a permission at all - it's power, power over others, not permission...
              Last edited by Libreman; 03 June 2012, 08:29 AM.

              Comment


              • #37
                copyleft vs. BSD license

                Permissive licenses have "5 freedoms":
                • The freedom to run the program, for any purpose (freedom 0).
                • The freedom to study how the program works, and change it so it does your computing as you wish (freedom 1). Access to the source code is NOT a precondition for this.
                • The freedom to redistribute copies so you can help your neighbor (freedom 2).
                • The freedom to distribute copies of your modified versions to others (freedom 3). By doing this you can give the whole community a chance to benefit from your changes. Access to the source code is NOT a precondition for this.
                • The freedom to distribute copies of your modified versions to others, under different conditions (freedom 4).


                Copyleft licenses are more free from software point of view. Permissive licenses are more free from user's/dev's point of view.

                Comment


                • #38
                  I am just waiting for one thing to happen. For GCC folks to rip off the "wonderful" LLVM + CLang compiler code and adapt it in GPL3+ GCC suite. That would be so ironic... Hehehe.
                  I can't stop giggling at thought of it.

                  Comment


                  • #39
                    That would be great for LLVM and Clang, but it won't happen.
                    Last edited by LightBit; 03 June 2012, 09:30 AM.

                    Comment


                    • #40
                      Why I would never use FreeBSD?

                      Why I would never use FreeBSD? I used it a bit, then I had chance to compare with Linux.
                      Basically, the conclusion is: FreeBSD guys do not care what would happen to their users at all.

                      1) Unlike Linux, they do not have package management system at all. No, really, they don't! So software management is a real headache compared to Linux. Sure, there are ports. However they're only convenient for "chosen few" and real nightmare for anyone else. To make things more funny, even some of these "chosen few" could be really scared by the idea to build whole OpenOffice or Firefox, etc.

                      2) Okay, if someone so inclined to building things from sources, you would expect at least convenient and trouble-free compilation, right? WRONG! By default you have a ancient GCC 4.2 which is bugged and seriously loses to recent GCCs like 4.6 in terms of code optimization. Then they plan to replace it with immature clang. So you have a very good choice. You can choose between gcc internal errors and clang internal errors. When you need to build 3rd party program not in ports, you will soon figure out that everyone on this planet has stored gcc 4.2 at the graveyard of technologies and never tests their program on this antuque stuff. Same problem with clang as well - nobody tests their programs on it, so it would be you who collects all internal compiler errors, etc. So you could easily end up working around compiler bugs, one way or another. Instead of just getting your program running! The most stupid thing is that it's 10 times easier to build same program even in binary-based Linux distro. They have a recent gcc and binutils and virtually all programmers are testing their software in similar setups so they will catch most of bugs and quirks instead of you. So unless program is in ports, building it from source could be a real nightmare.

                      3) The hardware support is bad-bad-bad. If you develop something for web, something for *nix-based systems, or simply prefer *nix-like approach, it can be far more convenient to have more or less the same system on desktop as well as on your target. However, FreeBSD can't really offer that. Most notably, they lack KMS so their Intel and AMD opensource drivers are from stone age and don't support recent hardware and have zounds of bugs, fixed years ago. You can also figure out that some wi-fi, controller or whatever does not works. If you have notebook, you will fugure out that a lot of things does not works at all. You maybe will be able to fix half of issues but why the ... on the Earth I should do that myself? Just because devs do not care what would happen to their OS users? I think it's a completely wrong approach.

                      4) Remembering about desktops and package systems once more, those guys haven't got one very simple idea: "one size does not fits all". Linux dudes learned it very well, so they conquered anything from small embedded to huge supercomputers. FreeBSD guys are slow. They still live in the world where nothing exists except their x86 servers. So they offer some "base system". Should I tell that I don't need bloatware like apache and bind on desktop? I don't need it on most of my web servers either. However, FreeBSD guys are fairly persistant on pushing their odd ways to the masses. Which seems to produce spectacular failure. As I seen few enterprises replacing FreeBSD with Linux for outlined reasons.

                      5) Well, they tell they're good for servers. But you see, they do not have virtualization support. In fact you can't use FreeBSD to host guest systems at all - it can't serve as self-sufficient hypervisor (like KVM) or as "hypervisor controller" (Xen Dom0). At very best it can be guest. But you would need some other kind of OS as hypervisor boilerplate. Managing 2 different kinds of OSes is more hassle than using one kind of OS as both hypervisor boilerplate and guests. To make things even worse for FreeBSD, on lightweight isolation front Linux guys have seriously outperformed FreeBSD "jails" with their LXC + Cgroups combo. Not to mention third party OpenVZ used by hosters almost decade or so.

                      6) To make things even more funny, FreeBSD can offer you either simple but antique, slow and clearly outdated UFS filesystem or fully blown ZFS enterprise monster, Which is terribly slow unless you have 64Gb RAM or more and performs well only on a limited range of workloads. To add up, ZFS seems to have stability issues on heavy workloads so when you complain about lockup, you're warmly welcomed with "are you using ZFS?!" questions. These guys absolutely do not have any medium-range solutions for file systems. You can choose between bicycle and rocket. But in their world there are no cars, bikes, airplanes and trains. This is very inconvenient to my taste.

                      7) There is virtually no distros who would at least try to fix at least some of idiocies of original system. For me it appears like if only apple and juniper are really interested in this system to use it as free boilerplate for their proprietary things without need to show their sources.

                      8) In fact FreeBSD is not crossplatform OS. You bound to x86 only. Formally they support ARM or MIPS but you would have a serious trouble to find hardware where it actually works to degree you can use it for something meaningful. Do you think BeagleBoard is looking cool? FreeBSD guys think otherwise. So while they have formal support for "ARM" you can't really have fun with some real piece of hardware to get it running with more or less all onboard peripherials alive and usable for some real-world tasks so you can have some fun (and maybe profit). You see, CPU without peripherials is like a plane without wings. These guys are way to slow to catch up with technologies and support SoC peripherials in a timely manner.

                      Bottom line: Linux appears to be far more promising and rewarding thing. Why would someone need a third rate solution plagued with technical and political problems when there is first-class solution for free and with a source code? OpenSource collaboration seems to clearly outperform old business model where free part attempts to pad intertests of proprietary corporations. Not to mention Linux guys are always making things running first, and doing politics, system polishing and brave declarations second. Unlike FBSD guys who seems to prefer quite an opposite approaches.
                      Last edited by 0xBADCODE; 03 June 2012, 09:48 AM.

                      Comment

                      Working...
                      X