*BSD News Article 3062


Return to BSD News archive

Path: sserve!manuel!munnari.oz.au!news.hawaii.edu!ames!olivea!uunet!mcsun!uknet!edcastle!aiai!jeff
From: jeff@aiai.ed.ac.uk (Jeff Dalton)
Newsgroups: alt.suit.att-bsdi,comp.unix.bsd
Subject: Re: UNIGRAM's article on the USL-BSDI suit
Message-ID: <7053@skye.ed.ac.uk>
Date: 4 Aug 92 15:51:37 GMT
References: <1992Aug3.143259.23897@crd.ge.com> <7045@skye.ed.ac.uk> <KANDALL.92Aug4161214@globalize.nsg.sgi.com>
Organization: AIAI, University of Edinburgh, Scotland
Lines: 94

In article <KANDALL.92Aug4161214@globalize.nsg.sgi.com>
kandall@nsg.sgi.com (Michael Kandall) writes:

>Jeff> If USL has any evidence that code was copied, they should be able
>Jeff> to say what code it was.  For some reason, they don't seem to want
>Jeff> to do this.  
>
>The issue may be more than copyright.

Fine.  I'd say the same thing (mutatis mutandis) about these
other issues.

>>  It seems to me that a lot of people want something for nothing, and
>>dislike AT&T/USL for trying to profit from their UNIX software. 
>
>Chris is right.  There is a wide-spread perception that ``UNIX'' --
>trademark, code, interfaces, ideas -- is public property.  And some
>of it (interfaces and ideas) may very well be.

A lot if it _is_ in effect public, because it's been published
in books (while not being patented), etc.

Nonetheless, I object to the implication that those of us who
think AT&T/USL is wrong "want something for nothing".  What we
want is for work done at Berkeley, made available by Berkeley,
and in part funded by public money (ie, we've paid for it via
taxes), to be available as intended.

>Jeff> I don't mind USL profiting.  What I mind is their trying to prevent
>Jeff> other people from using non-AT&T and non-USL code written at Berkeley
>Jeff> (and elsewhere).
>
>Its not just code.  Its an entire system: designs, specs and maybe
>parts of the implementation, lifted from USL property.

Again,  I'd say the same thing (mutatis mutandis) about these
other issues.

>Independent of the legal issues, it is clear to me that the BSDI
>people have taken their ideas from USL's UNIX System.  Where do you
>think they learned to write UNIX-like, mu, mt operating systems? 

What do you think happened at Berkeley during all the years of
work on Berkeley Unix (BSD)?  Just what ideas were taken from
USL?  You really ought to find out something about this before
reaching that sort of conclusion.

You also seem to be under the impression that there's something
mysterious or especialy difficult about writing UNIX-like, mu, mt
operating systems.  There isn't.

> They
>weren't born like that.  They looked at the real UNIX code, used the
>ideas, and wrote their own.  Whether the legal system combined with
>the USL Software Agreement protect those ideas is a different issue.

BTW, what do you think should be said of people who looked at
Minix and learned how to write  UNIX-like, mu, mt operating systems
that way?

>It is also contrary to the spirit of Open Systems.  As a company, what
>incentive do I have to invest person-years of time, money and
>collected expertise, designing and specifying a technology which I
>will license to other parties, if those other parties can:
>
>	license my stuff for a couple years
>	study my implementation and techniques
>	copy my ideas and implementation (changing the actual code)
>	cut me out of the picture

Again, you seem to have a completely mistaken idea of the history
of BSD Unix.

If the history really was as you seem to think, I'd agree with
you.

>Jeff> What's this about "rewriting individual routines"?
>Jeff> I hope you don't want to suggest that if someone has a copyrighted
>Jeff> (or otherwise protected) compression routine no one can write _any_
>Jeff> compression algorithm because that would be rewriting a routine.
>
>It was recently explained to me (I am no expert on this stuff) that
>this would be about software patents, not copyrights.  And, yes,
>patented techniques cannot be used freely.

Just so.  But the prohibition is not against rewriting individual
routines.  For instance, compression algorithms have been patented,
but that doesn't mean I can't write my own compression routine
(with, say, the same external interface).  I just have to use a 
different algorithm.

(Not that algorithms ought to be patentable.)

-- jd