Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!uwm.edu!cs.utexas.edu!uunet!van-bc!ubc-cs!manis
From: manis@cs.ubc.ca (Vincent Manis)
Newsgroups: gnu.misc.discuss
Subject: Re: Disinfecting the GNU Public Virus...er...License
Message-ID: <6055@ubc-cs.UUCP>
Date: 21 Dec 89 02:10:34 GMT
References: <4&VSZ:@splut.conmicro.com>
Sender: news@cs.ubc.ca
Reply-To: manis@cs.ubc.ca (Vincent Manis)
Organization: The Invisible City of Kitezh
Lines: 49

Somebody recently posted a 1-line article reading `Boy, is this
newsgroup stupid.' After going through 100 or so gnu.misc.discuss
articles, I'm inclined to agree with this person.

After reading the GPL very carefully, I am unable to see *anything*
different in kind from a standard licence agreement for a programming
language. Indeed, anything which incorporates FSF products becomes
subject to their licence agreement. This is also true of *every*
programming language product with which I am familiar.

For example: suppose that I buy Borland's optional Turbo C library
source package. I can link Turbo C library object modules into my
program, and then distribute the program without paying any royalties;
however, However, I can't make modifications in the library code and
then distribute the result as a library called `Manis C Library',
whether it be for money or not. (Borland, which has remarkably sensible
licencing policies, still retains ownership of the components of their
package.)

Similarly, Texas Instruments markets a quite good Scheme system, at a
very low cost. That system does not produce executables, so you need a
copy of (and hence a licence for) TI's Scheme system for each machine on
which you wish to run a program.

I have not heard any of the anti-FSF types in this newsgroup dumping on
Borland or Texas Instruments for their `restraint of trade' in refusing
to give their software away. Admittedly, FSF's licence agreement is
somewhat more restrictive than either Borland's or TI's, but, having
accepted the principle, all we're doing is haggling over where to draw
the line, as the Archbishop said to the chorusperson. 

It seems to me that FSF's sin, in the minds of its opponents, is to give
the code away, thus undercutting (in at least their own minds) some of
the readers' livelihoods.  But here's the paradox: if you believe that
this undercutting exists, then you have already conceded victory to RMS.
That's exactly his point in the GNU Manifesto.

What it comes down to is simple: deciding to use a particular software
package is a complex decision, based upon capabilities, performance,
cost, and licence conditions. In that regard, FSF is no different than
Ashton-Tate. (I happen to have a much higher regard for FSF's ethics
than for most software publishers, but that's a different thing.) If
Ashton-Tate can claim ownership of the dBASE IV code, why can't FSF do
the same thing with GNU C? 
--
\    Vincent Manis <manis@cs.ubc.ca>      "There is no law that vulgarity and
 \   Department of Computer Science      literary excellence cannot coexist."
 /\  University of British Columbia                        -- A. Trevor Hodge
/  \ Vancouver, BC, Canada V6T 1W5 (604) 228-2394