Iqigo Tejedor Arrondo wrote:
> Hello,
> 
> I have to do a clean install of a server, this weekend :)
> 
> I suppose that most suitable is -stable (3.9), and update when 4.0
> happen to -release... but I would like to hear opinions if somebody
> thinks different.

Given the choices you have right now, I'd probably go with 3.9 now, keep
the system as minimal as possible, then upgrade to 4.0 after it is
released...that is, if your hardware is fully supported.  It's pretty
painless to do, and it is good to get in the habit of doing it before
the system is relied upon continually.

If your hardware isn't sufficiently supported by 4.0, you have a bit of
a problem.  If you go with a snapshot, you are stuck to -current until
4.1 comes out (or stuck reloading and rebuilding from scratch on 4.0
later).  That's far from the end of the world, but it might be more
exciting than you are planning on.

Me?  My CDs arrived Monday.  Order early, order often! :)  (no, being on
the team doesn't get me CDs any earlier).

> P.D. like has been commented in another thread, I would like to read
> something about drivers in openbsd (howto, example, steps, or somehing
> like). And also I want to suggest: a list of btasksb for people who
> loves to collaborate, but not know to program/hack.

"howto"s for writing a driver sounds like it is a mindless formula; drop
in some manuals, a semi-warm body, turn a crank, and out pops a driver.
I don't think I'd ever short-change the OpenBSD driver writers like
that, it is not a formulaic process at all.

As for tasks...do something that needs to be done.  One good starting
place is:
   http://www.openbsd.org/want.html
Get hardware developers want/need in their hands, and magic happens.
Note the part about "hardware developers want/need"...  dumping junk on
'em, or dumping hardware where the issue is code doesn't help.  I
suspect plenty of developers have multi-proc SPARC machines...the issue
there is no one is writing the code, not the lack of hardware.  Again,
this is not a formulaic task, lots of hard and original work needs to be
done.

Which isn't to say that some people don't LIKE old junk...but you could
give me lots more SMP sparc machines, and OpenBSD/sparc SMP support
won't be a day closer.

Otherwise...just find something you don't like and fix/improve it.
Understand that the vast majority of "improvements" people come up with
aren't accepted as part of the base system, but absolute none of "just
talk" is ever accepted, and you will probably learn something in doing
the work...and learn more if it is rejected. :)

Nick.

Reply via email to