Slashdot Mirror
MicroBSD Is No More
TrumpetPower! writes "Recently there's been quite a row in the OpenBSD community over copyright infringement by the OpenBSD spinoff, MicroBSD. Many parts of MicroBSD would seem to be a wholesale search-n-replace of the two names...including copyright notices. As a result, MicroBSD has shut down. It's worth noting that, as of this story submission, the MicroBSD Web site is still up and running with no special notices."
Rather than bail out completely I'd like to see them become part of the OpenBSD project and provide a slimmed down OpenBSD option. It was extremely promising to have OpenBSD features on a device with only a minimal RAM disk. It was poor judgement to replace the copyright info. It's good they've fessed up. The idea of a microbsd is still a good one.
Hat's off to you. Winners never quit, and quitters never lose, but if you never win and never quit, you're just stupid
MicroBSD was an interesting project, and if we Slashdotters don't kill the server completely, it's interesting to look around. However, looking at the infighting and piss-taking on the mailing lists, it's still apparent that there BSD crowd makes problems for itselves.
Unfortunately there's too much of a "holier than thou" feeling amongst many BSD coders, but then again, getting rid of Matt Dillon (see earlier /. BSD story) won't help either. I want the BSDs to do well, but the community can be v. problematic...
Here's an example. After using Linux for several years, I gave FreeBSD a try and was very impressed by the solid kernel and coherent userland. There were a few things I couldn't fathom out though, like getting USB joysticks working, and asking on the mailing lists or similar forums always got the same kind of responses: "Go back to Linux if you want that", "FreeBSD is brill and doesn't need to support it" etc.
It's this zealotry, patronising attitude and belief that FreeBSD is the "one true OS" that is really damaging its acceptance. All OSes have their loudmouth advocates, but I've never seen anything like the hideous attitude that seems so common among FreeBSDers. And as said, it's all the more a shame because I was genuinely interested in the system and respected its good points.
And note that I specifically say FreeBSD - I've found that OpenBSD and NetBSD users realise that their OS isn't going to be ideal for all situations, and don't feel threatened by alternatives. It's the stuck-up FreeBSD zealots who think their OS is better than *everything* that are the real problem.
% man 1 banner
gives out: banner(1)
(can't paste it here, I get that lameness junk filter because of all those # characters).
I don't entirely blame them for closing down shop in this case. As a long-time proponent of Open Source I must admit it sometimes sickens me how annoyingly 'victimized' OSS authors can act when they sense a license violation.
I mean, sure, you *should* bring attention to license violations when you see them, but the OSS response tends to be way overboard, calling for boycotts, fatwahs, and whatever else before all the facts are known and before it is clear if the violation was really underhanded or just an oversight. It is this kind of religious zealot behavior that is holding OSS back from wider commercial adoption, IMO.
Whew -- for a minute I thought this was an article referring to a bizarre fallout with PicoBSD. PicoBSD is a neat little FreeBSD-on-a-single-floppy distro. Kind of an equivalent to Linux's admirable Leaf Project.
I hope GPL will be modified to require the source to be as easily available and featured as binaries and to be usable on it's own without any commercial software added on. Like Darwin distributions from Apple, not obscure changes that do not compile and do not explain what was done.
One of the distinctive things to note about Linux is that its code base is rather distributed. The kernel comes from here, cc comes from there, and the $other_thing comes from somewhere else. The boundary between the base Linux OS and additionally-installed software is sometimes not very defined. The BSDs' code bases are organized differently, each BSD having its own (mostly) centralized, integrated code repository and build system. BSDs do include some "contrib" software (e.g. Less or OpenSSH) that comes from other sources, but contrib releases are still merged and adapted into the main repository. BSDs have a very definite boundary between what is the base OS and what is a third-party package, or "port".
Either way is a great way to structure an OS code base. However, the organizational differences do have some effect upon what is the best way to make a small or embedded version of the OS. With Linux, it's good to fork off a separate distribution, so that packages and a build system can be most effectively engineered. Granted, I never looked at MicroBSD closely (so I could be totally wrong), but that seems to be the approach they took. With BSD, however, it just doesn't make sense to do it that way. If I just wanted to recompile the entire FreeBSD OS, I would do this:
Yadda yadda yadda, similarly in the other BSDs. To expand from this, I could build a fully functional embedded system with a measly few hundred lines of sh(1) script. Note how this did not require me to create another, entirely separate open-source OS project. I just piggybacked on BSD's existing, highly-integrated code base and build system.To conclude: making it easy to make a small BSD is at most a minor job, easily (and best) doable within the fold of one of the existing BSDs. Forking BSD doesn't make sense for this.
And before someone flames me, let me say this: I prefer BSD, but both Linux and BSD make great small or embedded OSes. Each has its own strengths and weaknesses. I don't wish to knock Linux; my only argument concerns the lack of necessity to fork BSD.
BTW, no puns intended by "forking BSD". :-)
Washington, DC: It's like Hollywood for ugly people.