Why FreeBSD not popular on hardware vendors

Chad Perrin perrin at apotheon.com
Fri Dec 12 10:12:12 PST 2008

On Thu, Dec 11, 2008 at 09:50:36PM -0800, prad wrote:
> On Thu, 11 Dec 2008 18:58:14 -0700
> Chad Perrin <perrin at apotheon.com> wrote:
> > So . . . are you saying that increased support for 3D accelerated
> > graphics is not an "improvement", and should therefore not be
> > considered a worthy goal?
> > 
> no. access to hardware probably is a worthy goal, however, you need
> people to write the software and it's up to the freebsd team(s) to
> determine if 3d graphics is or is not worthy, isn't it?

I don't recall anyone saying "I'm with such-and-such a FreeBSD
development team, and these are the reasons we aren't going to do
anything about that at this time:".  All I recall is several people
cropping up and saying the equivalent of "If we work on that stuff,
FreeBSD will just become MS Windows, and it'll suck."  I disagree with
that estimation -- but if someone wants to offer an actually reasonable
argument, I'm all ears (or eyes, since this is a textual discussion).

> > This is completely orthogonal to the question of whether people who
> > express a desire for better support for desktop functionality should
> > be excoriated publicly on this mailing list, and spanked for having
> > the audacity to want to migrate from MS Windows to FreeBSD for use as
> > a desktop OS.
> > 
> this is a pretty nice list and i haven't found much spanking going on
> here.

The "spanking" I have seen largely seems to focus on this particular
area, and is mostly championed by one person, though.  I guess I find it
even more offensive because it's an exception rather than the rule here,
and I rather like the otherwise helpful spirit of this community.

> > I agree that desktop usage should not take priority over more
> > fundamental quality concerns in FreeBSD development.  Telling people
> > to stick it in their ear when they say it would be nice to have Flash
> > support is not related to the ability to prioritize development
> > goals, though.
> > 
> i agree that telling people to "stick it in their ear" is not nice, but
> i don't recall anyone doing so. unfortunately, if i ask for evidence
> regarding this, you'll probably just tell me to RTFML as you did in
> your other reply.

It was a summary and paraphrase -- I don't recall anyone literally using
the phrase "stick it in your ear".  Please try to follow the discussion,
rather than being diverted by paraphrases, since I don't have the whole
mailing list archive memorized.

> > Desire for better desktop functionality doesn't have to equate to
> > wanting desktop-oriented development to "control the reins of
> > development" for the whole system.  Why the hell do you seem to think
> > it does?
> >
> i don't know why you think that's what i think. what i said was that
> was a concern. i certainly do know that in other areas
> (computer education for instance), user convenience has destroyed
> technical know-how (specifically, at some schools when the graphic
> interface emerged in the 80s, word-processing dominated programming and
> the some schools lost their thinkers). microsoft's catering to user
> desires has produced some rather inferior software too.

I think that's what you think because "control the reins of development"
was a verbatim quote of what *you* said.

I don't see greater core functionality and better driver support is just
superficial "user convenience".  It's not like I'm suggesting FreeBSD
should violate privilege separation so people don't have to worry about
the difference between user accounts and administrative accounts, or that
it should make booting into KDE without a password the default behavior
on boot so people don't have to worry about that icky CLI and memorize
passwords.  I'm not even suggesting that FreeBSD should adopt the MS
Windows default, automatic wireless network roaming behavior.

I'm just trying to suggest that opposition to discussing whether the
resources exist to address some driver issues is kind of silly (for

> may be it doesn't have to be that way, but often there is a price to be
> paid for 'convenience'.

There is, indeed, a price to be paid for (poorly planned) attempts to
improve convenience.  Luckily, that's not what I'm suggesting -- nor is
it what everybody else who would like an improved GUI environment is

> > Hell, I think the more server-oriented development
> > philosophy of FreeBSD is actually a big part of the reason it works
> > so well as a desktop OS! Maintaining a more server-oriented
> > development philosophy in *no way* precludes giving some attention to
> > strictly desktop-related functionality, though.
> >
> perhaps, but if you have a server-oriented philosophy, why would you
> give much attention to desktop-related functionality?

"More server-oriented" does not mean "exclusive of desktop".  It's not
like I said it should be "strictly, exclusively server-oriented, and
screw those people who use FreeBSD as a desktop system."

> i recall on the openbsd elist a couple of years ago people asking what
> wm is best. most of the answers went something like - the default twm
> (i think that's what it was) or fluxbox was "all i need". 

My preferences swing more toward AHWM and wmii, neither of which is
exactly a "bells and whistles" kind of window manager like Compiz Fusion.
That doesn't mean that functionality like strong support for 3D
accelerated graphics, which can improve the desktop user's experience, is

> > Pretending the two are incompatible goals, as a few notable people
> > here seem to want to do, is counterproductive in my opinion.
> >
> not necessarily. one group is saying we have a great os, so it would be
> even better if it could accommodate some of the fancy stuff that the
> kdes and gnomes etc offer even more. the other group is saying why
> bother, because who really needs it and if they want it they can get it
> elsewhere. i think the concern of the latter group is by no means
> illegitimate, because time and resources aren't unlimited.

On the other hand, their statements *do* imply that *my* position is
illegitimate in some way -- my position being that FreeBSD is the best
desktop OS I've ever used (in addition to being a great server OS), and
it could be even better if I didn't have to make a choice between either
screwing up the stability of my system or using substandard, crappy
drivers that emulate 1992 hardware capabilities for the ATI graphics
adapter in my laptop.

> on the otherhand, as i vaguely recall on a flash thread, someone said
> no one is stopping anyone from writing a better flash for freebsd if
> they really want to. i think it is ok to ask, but i don't think it is
> ok to expect. for me, freebsd is a gift and i don't have any
> expectations from those who put the effort and skill into creating any
> opensource initiative.

I'm using swfdec and MPlayer for Flash on my laptop.  I know there's
already effort being made in that area, and I'm glad -- and I also know
that, if I had the resources, I could contribute more to that effort.
Some people don't know that, and are basically told to go away by some
people when they bring it up.  Still other people suggest alternate
approaches to fixing the problem, and are also basically told to go away,
when a more appropriate response would be to say "I think you should talk
to the people at the swfdec and gnash projects about that," in most

Some of the responses seem to be equivalent to "Even if Adobe opened the
source for the Flash player plugin under the BSD license, and someone
fixed up all possible problems with the code, we still wouldn't want it."
I just don't understand that level of resistance at all.

Chad Perrin [ content licensed OWL: http://owl.apotheon.org ]
Quoth Coline MacDonald: "Don't burn your bridges while you're standing
on them."
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 195 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-questions/attachments/20081212/fd8d8d8f/attachment.pgp

More information about the freebsd-questions mailing list