*BSD News Article 34638


Return to BSD News archive

Newsgroups: comp.os.386bsd.development
Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!bunyip.cc.uq.oz.au!munnari.oz.au!spool.mu.edu!howland.reston.ans.net!europa.eng.gtefsd.com!MathWorks.Com!panix!ddsw1!a2i!dandelion.com!not-for-mail
From: Leonard N. Zubkoff <lnz@dandelion.com>
Subject: Re: Call for 386BSD Rel.1.0 SIG (Special Interest Group)
In-Reply-To: jkh@violet.berkeley.edu's message of 19 Aug 1994 03:45:54 GMT
Message-ID: <334kg1$1l2@kelewan.dandelion.com>
Sender: Leonard N. Zubkoff <lnz@dandelion.com>
Nntp-Posting-Host: dandelion.com
Organization: Dandelion Digital
References: <jmonroyCuFvqp.BKv@netcom.com> <3319ti$7rl@agate.berkeley.edu>
Date: Sat, 20 Aug 1994 10:04:49 GMT
Lines: 55

In article <3319ti$7rl@agate.berkeley.edu> jkh@violet.berkeley.edu (Jordan K. Hubbard) writes:

   Fine, we even believe you (sort of).  But then the question still remains:

   "Why follow Bill?"

No reason to at all.

   The other *BSD's are delivering solutions NOW, have delivered solutions for
   the last 18 months, during which time Bill had abandoned his users to an
   uncertain fate, and show every sign of continuing to do so.  Considering
   that Bill's track record, not to put too fine a point on it, SUCKS, why
   should we trust him?

I was somewhat disappointed in that I went to the talk hoping it would be far
more technical in nature.  What it provided was mostly an overview of the
directions Bill's work had gone since the last talk about 1.5 years ago.

I went both this week's and in Spring '93 to the SVnet talks Bill gave, and
came away from it with a rather different view than perhaps some others have.
After the most recent talk, I'm even more convinced that 386BSD is not about
delivering product solutions of any kind.  Rather, it is about delivering
operating system research and explanations, essentially a detailed book about
OS design that happens to come with runnable code.  My impression is that the
primary purpose of the delivered source code is to have working code rather
than pseudo code to write about and explain, so that people can examine the
code directly and modify it to test out other algorithms.  More than once he
spoke about the need for the reasons behind all the design decisions to be
clearly presented, as a legacy for future researchers, so that the heart of
Berkeley Unix would not be lost.  This is a valuable contribution, it's just
not the one some people were hoping for.

I really never got the impression from this week's talk that the fundamental
purpose was to deliver anything resembling a production quality operating
system.  Thinking back on the Spring '93 talk, it had very much the flavor of
talking about research directions as well, though it was easy to infer a
product quality to it, probably because that's what *I* wanted to believe.

Well folks, the product is probably not what we've come to want from our
NetBSD, FreeBSD, or Linux worlds -- the Jolitz' fundamental product seems to be
their writings, and one of their goals to make those writings available to a
wide range of people.  Just look at the manner in which they are making it
available -- accessibility in Windows sounded far more important than in its
native environment!

I may have this wrong since I wasn't paying attention during the early days of
386BSD; I am only going on what I heard during these two talks.  Other folks
who are closer to 386BSD may have seen something quite different, and I
encourage them to explain their perceptions of the talks.  However, if my
understanding is correct, it would go a long way towards explaining why we have
the separate groups working on *BSD systems we have today.

Enough rambling for now...

		Leonard