UNIX admin wrote:
We do need more people, but not the Linux hacker kind. All those would want to
do is muck with Solaris so that it looks, works and behaves like Linux.
Unfortunately, Linux suffers from serious lack of engineering and quality
control because everything is implemented ad-hoc by people who think they know
UNIX better than profesional engineers and scientists that have spent the
better 35 years studying and working on computing challenges and problems.
Believe me, that is not what you want for Solaris; the quality would suffer,
and Solaris would no longer be what it is, and what it makes Solaris so great.
I believe that the kind of people Solaris would most benefit from are the BSD old
skoolers. Those guys actually *care* about the quality of their work, even if they might
not necessary always reach it. At the very least, they *strive*, instead of the "it
works for me, that's all I care about, if it doesn't work for you, you have the source
code" Linux hacker attitude; not everyone is a UNIX programming guru; some people
just want to *use* UNIX and get *stuff done* on UNIX. Not everybody is a programmer, and
assuming that everybody would have to be is just plain wrong ideology. That's why our
GNU/GPL/Linux pals don't really have any long term future, no matter how many *millions*
of hackers they have at their disposal.
Being a BSD somewhat-old skooler (but not old-old skooler, I wasn't in
CSRG ;-)), I can say that in the *BSD world, there has always been great
respect for Solaris. It has always been seen as the OS that got a great
number of things right, and often, during technical discussions, at
least one person always came up with "let's see what Solaris did"
(though we couldn't look at source code).
To your more general point: growing an open source OS community is
somewhat of a balancing act. Putting an emphasis on code quality and
"educating" your developers is good. But, you also have to promote
yourself, otherwise you won't attract any new developers in the long
run. With popularity also come annoyances. There will be much more noise
on mailing lists, which can drown out the good technical discussions,
leaving the early developers longing for the "old days". But, without
popularity, you'll be destined to occupy no more than a niche. NetBSD is
an example of this. We always stressed code quality and portability, but
neglected features that attract average users, and neglected promoting
ourselves. Which led to NetBSD becoming probably the smallest *BSD
community (though it's not easy to have hard numbers on that), despite
being the first of the current *BSDs to come into existence.
That's why I'm glad that there is a push to expand the OpenSolaris
community from within Sun. We need to build aa fertile base to grow
developers from, so to speak. However, I have strong doubts whether
adding GPLv3 as a license is a good way to do this.
- Frank
_______________________________________________
opensolaris-discuss mailing list
[email protected]
