*BSD News Article 3666


Return to BSD News archive

Path: sserve!manuel!munnari.oz.au!mips!mips!sdd.hp.com!usc!news
From: merlin@neuro.usc.edu (merlin)
Newsgroups: comp.unix.bsd
Subject: Re: Restrictions on 'free' UNIX / 386BSD (Re: selling 386BSD)
Date: 14 Aug 1992 18:44:16 -0700
Organization: University of Southern California, Los Angeles, CA
Lines: 37
Sender: merlin@neuro.usc.edu (merlin)
Message-ID: <l8oobgINN47m@neuro.usc.edu>
References: <x> <l8n8qcINN2c5@neuro.usc.edu> <1992Aug14.205511.15778@craycos.com>
NNTP-Posting-Host: neuro.usc.edu

In article <...> jrbd@craycos.com (James Davies) writes:

>Excuse my ignorance, but doesn't 386BSD use gcc?  Sounds like it's already
>contaminated beyond all hope of redemption...

You are of course not ignorant -- nor is anyone else I have met on the net --
gcc is an isolatable special case component clearly identified as a special
case in the distribution -- it is not a simple driver -- nor is it a major
patch -- moreover no fee is required of anyone redistributing gcc -- nor does
the gnu organization require specific permission for redistribution.  There
exist portions of the gnu license to which I object -- but since I am at all
concerned about doing my own compiler development the restrictions have very
little impact on me.  This is very different from vague intellectual property
claims (where no one knows the terms for free redistribution) or the specific
restriction (every penny of any payments to jolitz [without even a reduction 
to cover media, machine time, or personnel time expended in the duplication]).

Moreover, gcc could be stripped out without affecting many end user sites --
they just have to acquire compatible adaptors and put them at good addresses.
Including gcc in the present distribution doesn't make much difference -- but
a whole slew of cgd type copyrights and restrictions would be unmanagable.

For the record, I received a note from Bill Jolitz indicating (1) cgd should
not be singled out for criticism [I agree -- his was only a convenient case]
and (2) my point was right on the mark -- this is why Jolitz continued the
minimal requirements of the UC Regents distribution.  The net may not be at
all ready for this concept -- 'freely redistributable software' -- but there
are forward looking people like Jolitz who perceive the folly of inhibitory
and/or confusing ad hoc collections of miscellanous copyright claims.

Merlin

The Magician
King Arthur's Court
Camelot