What to expect in OpenBSD 5.0 onwards | Hacker News Reader

What to expect in OpenBSD 5.0 onwards

bsdly.blogspot.com

29 pointsyankcrime8 comments

What I personally think is interesting is that, as close as we are to OpenBSD 5.0, there are still no patches for 4.9. I don't recall a release that went without patches.

As a BSD fan in general, I do wish there was a bit less fragmentation in the BSD community. With the dwindling user community of OpenBSD, and also FreeBSD to a degree, a part of me wishes these teams would join forces to a bigger extent than they currently do. I don't necessarily mean that there should be one "BSD" distribution, but rather shared codebases and such, making it easier to take feature X from FreeBSD and using it in OpenBSD, and vice versa. Whenever a technology (such as ZFS) was implemented in FreeBSD, it should be extremely easy to port to OpenBSD (on that note, there may be licensing issues that keep OpenBSD from using ZFS).

Ex-FreeBSDer who's thinking to go back to twiddling with it again here!

I agree. These people should join hand to move *BSD forward faster. I feel that as days go by, BSD is losing its ground against Linux (and Windows).

It's also nice to have one shared codebases for the OS level stuff (sure, the GNU toolchain part can be different) so more people can actually learn them as opposed to be given a hard choice of one of the BSDs.

>BSD is losing its ground against Linux (and Windows).

why is that important? why do the BSDs have to be gaining on Linux to be a viable, useful system. If your entire goal becomes "competing" with Linux, you miss the entire point many people chose a BSD to begin with, a stable, well thought out, usable system. The ZFS integration clusterfuck with freeBSD pretty clearly shows the results of trying to rush things so they are somehow better than linux.

Merging all the BSDs together to "fight" Linux is stupid and shortsighted, the OpenBSD philosophy is about more than security tweaked defaults for freebsd, its an entire process from conception of a new program, to implementation, to maintenance of the code. If you try to reduce it to a common codebase, it is just going to be freebsd + tweaks, which negates the entire reason people want to use OpenBSD to begin with.

You might be right. The problem with software I suppose: if it grows, there will be pain. If it doesn't, well, let's just hope it won't stall and die quietly.

I don't mean to compete or fight against Linux in that sense. I suppose I just want to see more things happening in BSD land (Amazon instances, VirtualBox, native Java) but if that happens, could potentially disturb the stability of *BSD itself.

And yes, I do know (a bit) the principal behind OpenBSD that may not work well with other BSDs. Different goals that lead to different software development process.

This already happens to a greater extent than maybe you realise. Witness FreeBSD's adoption of OpenBSD's pf, and so on. Each of the three 'main' BSDs (Free, Net, Open) have a very different focus in terms of the direction of each project, and these idealistic differences - especially when you throw people like Theo de Raadt into the mix - will mean that it's always going to be fragmented in such a manner.

I kind of like it that way, personally.

Yeah, I like it that way as well. Sadly, this means that you'll always miss some killer features that are available in other BSDs (FreeBSD's jails, NetBSD's puffs, DragonFly's HAMMERFS).

But if it would be extremely easy to port kernel-level features from one BSD to another, then it would mean that their code doesn't differ too much and there would be no point in having multiple BSDs in the first place...

Reply on news.ycombinator.com