I identify with what Poul-Henning Kamp says in many ways. As most of us here probably fall into the younger category, I can definitely see the value in a more carefully designed and engineered experience. That is one of the things that draws me to OpenBSD. I also like FreeBSD but really only because of a few features OpenBSD lacks (ZFS primarily). When something is bloated, has lots of problems, and is unsatisfactory, the OpenBSD developers write a replacements that works so much better. I really like that!
I am perfectly content carefully configuring my OpenBSD system to be exactly the way I like it. I used to try and do the same with Linux. I started with RedHat back in the nineties but very quickly moved to OpenBSD. I tried again with Gentoo and Crux but just couldn’t get the simplicity and excellent base system design that I liked in OpenBSD. I have also used Debian for a significant amount of time but I always felt like it was a huge, slow-moving operating system target that would get there long after everything else (it’s just my feeling, not necessarily a fact).
In the last few years I have started using Ubuntu and it fits a different paradigm for me. I am also primarily an OS X user on the desktop. Ubuntu fits this same sort of paradigm for me. OS X has lots of great applications available that are both commercial and open source but I rarely compile anything from source (I use MacPorts for tmux and a couple of other things). Ubuntu falls into this same paradigm. I can download all sorts of pre-compiled packages for Ubuntu that fit my needs for most things well but I have deliberately used a different mindset of a desktop environment where I can just get things done and don’t really enjoy the system.
If I could pick all types of work I have to do, I would choose to use OpenBSD exclusively. But, irritatingly, that’s not possible at this time. The potential barriers to this are much less than they ever have been. Maybe I will still get there.
In a sense, you could argue that iOS (and OS X) is more of the cathedral model in a commercial sense. Apple controls it and has set forth pretty careful engineering guidelines that must be met or a potential app will be rejected. All of this is not an open model in any sense of the word but has made a major commercial ecosystem and created many jobs for app developers. I like this controlled environment because you get exactly what you expect on every device running iOS 5, for example.
There really are a variety of ways to do things. I happen to like iOS and OS X quite a bit but nothing compares to OpenBSD for me in terms of design and engineering. No system is perfect but I do like having choices.
I’m a very strong supporter of open source software (in particular, ISC-licensed and public domain) but the fact of the matter is there is no quality assurance in open source software. In theory, there is; but it is far easier to just put up with the glitches than to send in a patch.
Agreed. That’s really why I like the OpenBSD project. If you write terrible code and submit it, you’ll probably get chastised. In my opinion, it has the best quality assurance of any open source operating system I know of. It’s hard to move from the new developer phase (me) to putting out great code but a large body of well audited code like OpenBSD is a great place to learn.
Absolutely. I’m also a fan of when companies develop software, but make the source publicly available. You pay them to fix it and implement features, and if you need to verify things (or want something fixed faster) you can submit code. Not terribly common in practise, though, unfortunately.
I identify with what Poul-Henning Kamp says in many ways. As most of us here probably fall into the younger category, I can definitely see the value in a more carefully designed and engineered experience. That is one of the things that draws me to OpenBSD. I also like FreeBSD but really only because of a few features OpenBSD lacks (ZFS primarily). When something is bloated, has lots of problems, and is unsatisfactory, the OpenBSD developers write a replacements that works so much better. I really like that!
I am perfectly content carefully configuring my OpenBSD system to be exactly the way I like it. I used to try and do the same with Linux. I started with RedHat back in the nineties but very quickly moved to OpenBSD. I tried again with Gentoo and Crux but just couldn’t get the simplicity and excellent base system design that I liked in OpenBSD. I have also used Debian for a significant amount of time but I always felt like it was a huge, slow-moving operating system target that would get there long after everything else (it’s just my feeling, not necessarily a fact).
In the last few years I have started using Ubuntu and it fits a different paradigm for me. I am also primarily an OS X user on the desktop. Ubuntu fits this same sort of paradigm for me. OS X has lots of great applications available that are both commercial and open source but I rarely compile anything from source (I use MacPorts for tmux and a couple of other things). Ubuntu falls into this same paradigm. I can download all sorts of pre-compiled packages for Ubuntu that fit my needs for most things well but I have deliberately used a different mindset of a desktop environment where I can just get things done and don’t really enjoy the system.
If I could pick all types of work I have to do, I would choose to use OpenBSD exclusively. But, irritatingly, that’s not possible at this time. The potential barriers to this are much less than they ever have been. Maybe I will still get there.
In a sense, you could argue that iOS (and OS X) is more of the cathedral model in a commercial sense. Apple controls it and has set forth pretty careful engineering guidelines that must be met or a potential app will be rejected. All of this is not an open model in any sense of the word but has made a major commercial ecosystem and created many jobs for app developers. I like this controlled environment because you get exactly what you expect on every device running iOS 5, for example.
There really are a variety of ways to do things. I happen to like iOS and OS X quite a bit but nothing compares to OpenBSD for me in terms of design and engineering. No system is perfect but I do like having choices.
the “hack up or put up” mentality is what sucked me in the most. :D
I’m a very strong supporter of open source software (in particular, ISC-licensed and public domain) but the fact of the matter is there is no quality assurance in open source software. In theory, there is; but it is far easier to just put up with the glitches than to send in a patch.
Agreed. That’s really why I like the OpenBSD project. If you write terrible code and submit it, you’ll probably get chastised. In my opinion, it has the best quality assurance of any open source operating system I know of. It’s hard to move from the new developer phase (me) to putting out great code but a large body of well audited code like OpenBSD is a great place to learn.
Absolutely. I’m also a fan of when companies develop software, but make the source publicly available. You pay them to fix it and implement features, and if you need to verify things (or want something fixed faster) you can submit code. Not terribly common in practise, though, unfortunately.
That’s a great model and I wish it existed more often.