Path: utzoo!utgpu!news-server.csri.toronto.edu!clyde.concordia.ca!uunet!snorkelwacker!tut.cis.ohio-state.edu!pt.cs.cmu.edu!dsl.pitt.edu!dsl.pitt.edu!sean
From: sean@dsl.pitt.edu (Sean McLinden)
Newsgroups: comp.windows.x
Subject: Re: You get what you pay for (was: Re: Motif -> Open Look look & feel)
Message-ID: <1990Jul23.142314.6541@cadre.dsl.pitt.edu>
Date: 23 Jul 90 14:23:14 GMT
References: <9007210429.AA09366@Larry.McRCIM.McGill.EDU> <3725@auspex.auspex.com>
Sender: news@cadre.dsl.pitt.edu (Usenet News System)
Organization: Decision Systems Laboratory, Univ. of Pittsburgh, PA.
Lines: 48

In article <3725@auspex.auspex.com> guy@auspex.auspex.com (Guy Harris) writes:
>>The UNIX that caught on like wildfire, that made it the success it is,
>>*was* source-available essentially free.
>
>To *some* customers.  Commercial source licenses for V6, for example,
>were somewhere in the $20K US range - or does costing $20K US count as
>"essentially free"?
>
>The "UNIX that caught on like wildfire, that made it the success it is"
>was V7; that was the first one that appeared on *lots* of boxes.  I
>suspect the availability of cheap *binary* sublicenses had more to do
>with that than the availability of cheap source licenses.... 

Perhaps. But it was the easy availability of Unix in academic environments
and subsequent exposure of a large number of programmers to Unix systems
via the Berkeley distribution (which was source), that created the demand.
That, coupled with the demonstrations by Sun and Apollo that Unix could be
a viable commercial operating system (even if you did need to have a systems
programming staff to keep it running), is probably what forced the rest of
the industry to concede. ATT didn't start charging standard commercial rates
for Unix until well after BSD hit the campuses and I think what hurt, there,
was not so much the price (have you tried to get a source VMS license?),
but the cost for derivative distributions. In fact, although I consider
ATT's licensing policies to be reasonably competitive with the industry
(for Unix, anyway), I think that their approach was errant for philosophical
rather than financial reasons and is what led to proliferation of the
two Unixes.

Another interesting thing about the Berkeley distribution was the number
of people who became, essentially, expert systems and applications programmers
in the absence of nothing other than the standard set of ATT->Berkeley
reference manuals, an article in the Bell Systems Technical Journal, and
Kernighan and Ritchie. On our own campus, among the C/Unix programmers,
there was probably accrued over 300 man-years experience with Unix before
the first formal course in Unix or C programming. The availability of the
network contributed, substantially, to this, I am convinced.

This is, of course, highly speculative. It would make for an interesting
social study (perhaps NSF would fund it), for someone to seriously examine
what the social and commercial effect was of that first little DARPA grant
to Berkeley to develop the BSD distribution. The knowledge from such a
study might be quite valuable to the industry and would (IMHO) cause a
lot of people to rethink what their roles should be in the commercial side
of this industry.

Sean McLinden
Decision Systems Laboratory
University of Pittsburgh Medical Center