*BSD News Article 42875


Return to BSD News archive

Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!msunews!gmi!zombie.ncsc.mil!news.mathworks.com!news.duke.edu!godot.cc.duq.edu!hudson.lm.com!netline-fddi.jpl.nasa.gov!elroy.jpl.nasa.gov!delphi.cs.ucla.edu!oahu.cs.ucla.edu!scottm
From: scottm@oahu.cs.ucla.edu (Scott Michel)
Newsgroups: comp.unix.bsd
Subject: Re: Linux vs. BSD?!
Date: 24 Feb 1995 23:49:40 GMT
Organization: UCLA Computer Science Dept.
Lines: 68
Message-ID: <3ilrak$hrh@delphi.cs.ucla.edu>
References: <3i7ar8$ahv@marton.hsr.no> <3i9aa3$sbp@fido.asd.sgi.com> <3iae19$8do@agate.berkeley.edu> <3iakqv$aj5@fido.asd.sgi.com> <3icfuv$amb@agate.berkeley.edu> <3idte4$olb@fido.asd.sgi.com>
NNTP-Posting-Host: oahu.cs.ucla.edu
X-Newsreader: TIN [version 1.2 PL2]

Larry: You need to calm down and write complete sentences. There are places
in your don't amek snece. :-)

Larry McVoy (lm@slovax.engr.sgi.com) wrote:
: Point #1: any OS that does not commercial support isn't
: interesting to me.  If you don't want DhFreeBSD to be included as a
: candidate for commercial supprt, then I could care less how good or bad
: it is.

But then again, neither does Linux. Creating and cutting distributions
doesn't constitute commercial support. There also has to be incentive
for commercial support, which currently doesn't exist, unless you can
in fact capture Linus and impress him into service.

: Point #2: any OS that gets commercial support will be perceived, by the
: MBAs that work at the company providing that support, it will be perceived
: as having intrinsic value.

I think what you meant that commercial support creates the illusion of
stability which MBAs see as a *good thing*. Yes, generally, this is true.
Trying to take on OS support on one's own is counterproductive.

: Point #4: I want to work for a company that does the best job of providing
: a commercially support OS and I want that OS to be copyrighted in such a
: way that the company can never lock it up - they have to "own" it by
: "owning" the people that know the rules of caring and feeding the OS.

Indeed, I can just see SGI doing this. :-) And I say that they pretty
much own you too. :-)

Basically, you'd like a cradle to grave employment situation. There are
some situations where sourceware is appropriate, and maybe it is
appropriate to OSs. Sure, we all get pissed off at OS vendors, mine happens
to be SCO. Like many developers, I get pissed with the compiler doesn't
work or a system call comes back funny. It creates a burden on supporting
ports.

However, taking your comment further, let's look at Cygnus. Cygnus works
because they have support contracts with companies that need support. It
pays the bills. Do they release all of the code they develop under contract?
No. They don't. So the copyleft has been end-run. Then let's look at where
Cygnus fits into the market. They are a niche player of cross platform
tools. There are other niche players. Did Cygnus become successful because
the software they maintain was and is still free? Partially. Surely free
software cut their initial development time to zero, but there is an
inherent defect between supporting the software and making the source code
available on demand (and essentially to anybody). Cygnus could easily abide
by the GPL and not release it changes to the public. And charge for the
changes and upgrades, like an evil, assinine, greedy MBA might.

: Jordan, Jordan, Jordan.  You think that would happen if there wasn't a big
: copyleft stick to force them to do it?  Bzzzt, thank you for playing, next
: contestent.  :-)  Really, it doesn't work that way.

If I could find a way of end-running the copyleft, I would. I don't think
a copyleft woud ever stand up in court with the additional restrictions
it makes. It is an intrusive document that violates one's individual rights
in the process of attempting to protect the Gestalt's. Remember "To protect
your rights, we have to take some of your's away."?

Jordan is right: The technology transfer process has to be renegotiated.
R&D is being consolidated to the university level. Whether or not the source
is available is a moot strawman. What matters is whether or not a viable
commercial interest can be built to market and support an viable user base.

I'd be more than happy to head one up when I finish my MS.

-scottm