Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watnot!watmath!clyde!rutgers!mit-eddie!canna.kr!rms
From: rms@canna.kr.UUCP
Newsgroups: comp.emacs
Subject: "Public" funds a la Wegrzyn
Message-ID: <8704010458.AA06172@canna.kaist.ac.kr>
Date: Tue, 31-Mar-87 23:58:11 EST
Article-I.D.: canna.8704010458.AA06172
Posted: Tue Mar 31 23:58:11 1987
Date-Received: Thu, 2-Apr-87 03:12:51 EST
Sender: daemon@mit-eddie.MIT.EDU
Reply-To: rms@PREP.AI.MIT.EDU
Lines: 29

It is true that the funding from the president of Ann Arbor Terminals
constitutes a sort of public trust.  Most software developed nowadays
with public funds is proprietary (you must know of dozens of
proprietary programs developed by tax-exempt universities), but I
consider that to be stealing from the treasury.

The GNU copyleft does a better job of defending the freeness
of a program than public domain status does.  The public domain
was designed for literature and does not work for software.
I think I am upholding the public trust well, distributing
GNU CC as I am.

The public trust aside, we don't have any private obligations in this
matter.  As far as Len Tower and I can recall, the president of Ann
Arbor Terminals never made public domain distribution a condition of
the funding.

This shows that we are behaving as good citizens and honest ones.
A completely unrelated question that you might want to ask is whether
we are obeying the law.  I believe we are.  In the absence of any
agreement signed by Len Tower, he is not legally restricted.  But even
if the portion of the compiler he originally wrote had been officially
a project of the foundation of the president of Ann Arbor Terminals,
and had been placed in the public domain, the law would permit us to
include it in a copylefted work, just as it would permit AT&T to
include it in a copyrighted proprietary product.

And this illustrates my first point: that a copyleft upholds the
public trust better than the public domain does.