"The users dilemma" (Re: "Free software" vs. "Open Source" (again))

[Eric Dunbar]

On 6/18/05, Eric Feliksik <milouny at gmx.net> wrote:
> Emil Oppeln-Bronikowski wrote:
> > On Friday 17 of June 2005 10:15, Eric Feliksik wrote:
> >
> >>When the discussion turns to free vs. non-free drivers, I think that
> >>Theo de Raadt's (from OpenBSD) work for free drivers [1] is a good
> >>thing.
> >
> >  Well, Theo may be a good programmer, but he's an rude troll, too. We're
> > losers[1] , because of system we use.
> >
> > [1] http://42.pl/url/cWp
> 
> Yes, he is rude, and I think he really lacks some social skills. My
> point is that he stands for software freedom and I think that is a good
> thing. I appreciate his work for free drivers, too.
> 
> No, as far as I can see he did not say that. That title is chosen by the
> editor/journalist.

Looks like it's a skillfully written piece of journalism, designed to
drive eyeballs to their adverts.

As for the comments attributed to Theo de Raad, well, it's hard to
judge what exactly are his opinions. I certainly would *not* call him
a troll. He's merely expressed an opinion (a widely held one at that)
seemingly rooted in and informed by experience, and, we're only seeing
a *tiny* fragment of what was likely a much larger conversation.

Quotes taken out of context can be extremely misleading (we recently
had a fake scandal involving a federal minister here in Canada b/c an
opposition right-wing member of Parliament (MP) taped him and
"accidentally" left out sections of the discussion in the first
version of the tapes he released which exhonorated the minister).

Linux *is* a hodge podge of hacks and they are by no stretch of the
imagination all good hacks by-and-large (by definition a hack is not
good since it's unplanned). The only reason that Linux is a success is
its development model -- there are so many beta testers that bugs are
"organically" fixed through the iterative process. It doesn't
necessarily mean that the code produced is GOOD or efficient, merely
that it is relatively bug-free (not the same as good or efficient...
proper planning can produce tight code with bugs but that performs far
more efficiently).

Find bug. Fix it. Create new bug. Fix it. Find old bug. Fix it again.
Realise a road map is needed for development. Hack one together. Find
new bug fix it. Forget about road map. Fix an unknown bug. Add
feature. Fix it. Throw out feature. Realise that if you'd stuck to
road map in the first place bugs 3, 5 and 8 wouldn't have crept into
the app ... you get the picture ;-)

I'm sure the quality of the projects does vary by the experience and
competency of the participants. Hopefully the best coders are working
on things like X, GNOME, and OO.org and all the way down the line
until you get the least experienced coders working on games and other
such non-critical components.

The fact that Apple chose BSD speaks volumes to the quality, maturity
and vision of that OS (granted, BSD also has more favourable licencing
terms but Apple did more with BSD in the three years it took them to
get OS X 10.0 to market than Linux has done since inception).

Apple committed to a full-blown desktop and server OS strategy that
revolves around *nix. The other "big" players merely dabble in *nix
and have shown little committment to *nix. As such Linux doesn't have
to do much well or have a vision and "will do". It's good enough for
what they want (servers, distributed computing), but, since they're
not going head-to-head with Microsoft they don't have to worry about
overall quality and long-term development goals.

At present, only minor players are using Linux (and X) for any serious
work. In time Linux will get better and larger market share, but, as
Microsoft showed so well in the 1980s and early 1990s, quality
software is NOT is _any_ way, shape or form linked to market place
success. Marketing (and attitude) _IS_. Success is not necessarily
equal to quality!!!

Besides, as for the I'm a Linux user and she's a BSD user dichotomy --
it's a false one. If you value the principles of FLOSS then it
shouldn't matter to one whether BSD or Linux "wins". As long as FLOSS
in general improves is all that should matter. And, if BSD has a
stronger development model then that is the model that should be
supported and supported for that reason. Linux should not receive
support *because* it's common but because it's the better model. If
it's not it should be abandoned in favour of BSD (though, as is fairly
widely known, the underpinnings of much of the low-level Linux *are*
BSD... if that tells you anything about BSD ;-).

Eric.