Future of FreeBSD

[Kevin Oberman]

> Date: Fri, 9 Apr 2004 12:14:07 -0700 (PDT)
> From: Nathan Seven <scosol at yahoo.com>
> Sender: owner-freebsd-current at freebsd.org
> 
> 
> --- Nathan Seven <scosol at yahoo.com> wrote:
> > >-- Arjan van Leeuwen <avleeuwen at piwebs.com> wrote:
> > > On Friday 09 April 2004 00:45, Atte Peltomaki
> > wrote:
> > > > Hello FreeBSD users and developers!
> > > >
> > > > As an active FreeBSD user, I'm ever so
> > interested
> > > about the future
> > > > plans of FreeBSD and direction of developement.
> > > Many of the features
> > > > that 5.x taunts are very impressive. But as of
> > > late I have been
> > > > increasingly worried about the direction (or,
> > lack
> > > of direction) things
> > > > have been going.
> > 
> > > This looks like a troll, and if it isn't, it
> > doesn't
> > > belong on this mailing 
> > > list. Please don't feed the trolls.
> > 
> > Hmmm- troll or no, I don't see it as a particularly
> > invalid question...
> > I have some of the same concerns myself-
> > Is there a specific "guidance" or "direction" list
> > that I should be subscribed to?
> > If not, this *would* be the proper place- no?
> 
> Errr nevermind- I thought that the above was the
> entirety of the "troll" post ( I saw the rest- troll
> indeed).
> 
> Anyway, I mentioned *I* had some of the same
> questions, I feel like I should clarify them-
> My questions have been in the management of the 5.x
> RELEASES- it's been quite some time since 5.0, and the
> 5.x series is still touted as "scary experimental
> stuff that will break"-
> I just think that's really hurt the adoption of it-
> For basic serverside stuff, bits like ACPI and sound
> aren't needed at all-
> Perhaps once the scheduling and pthread stuff has been
> solidified, a "stable base" release should be made?
> Meaning that the release as a whole should still be
> considered unstable, but using the as-shipped "stable"
> kernel config, things should be nice and solid?
> 
> Anyway, just my 2 cents- I don't need a response or
> anything, just making my thoguhts known.

On the vast majority of systems, V5 is very stable, but there are still
too many problems to declare it STABLE. 

The assumption that servers don't need ACPI is simply untrue. More and
more newer BIOSes simply will not boot without ACPI. Since ACPI takes
over the functions of APM, too often people assume that it does nothing
else, when it may well do a great many more things, many well outside
the bounds of power management and essential to system operation. And
ACPI support is now pretty good, especially for servers. Laptops still
see an assortment of issues with suspend/resume and PCMCIA, although
many of these have been resolved recently and some older boxes either
lack ACPI or have one so broken that it is useless. Most of those do
fine with nothing or APM, though.

The new thread libraries are getting very good and I have not used the
old library for some time and have had no problems as a result,
although the transition was a bit rough. I understand that others have
had more problems than I have had, though. I only have three system
running CURRENT, 2 desktops and my laptop.

The number of things needed before V5 is ready to be given a STABLE
label is now getting small, especially on UP i386 platforms. There are
more issues on other equipment, though. I suspect 5-STABLE will arrive
before the end of summer and, perhaps, long before. I'm not a part of
the release process, but I see no reason for a "stable base" release or
anything but a true stable.
-- 
R. Kevin Oberman, Network Engineer
Energy Sciences Network (ESnet)
Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab)
E-mail: oberman at es.net			Phone: +1 510 486-8634