Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!csd4.milw.wisc.edu!bionet!agate!eris.berkeley.edu!mwm
From: mwm@eris.berkeley.edu (Mike (I'll think of something yet) Meyer)
Newsgroups: comp.lang.misc
Subject: Re: The GNU license.
Message-ID: <26882@agate.BERKELEY.EDU>
Date: 2 Aug 89 07:26:31 GMT
References: <10596@riks.csl.sony.JUNET>
Sender: usenet@agate.BERKELEY.EDU
Reply-To: mwm@eris.berkeley.edu (Mike (I'll think of something yet) Meyer)
Organization: Missionaria Phonibalonica
Lines: 46

In article <10596@riks.csl.sony.JUNET> diamond@csl.sony.co.jp (Norman Diamond) writes:
<There is only one case where the copyleft provision is a problem.

And that case doesn't really exist.

<If someone wants to produce truly public domain software but does not
<volunteer the effort to keep providing copies of source [even if he
<can be paid for volunteering that effort], it seems to me that if such
<a person uses free tools, he is still being ethical.  Unfortunately
<copylefted tools are not free in this case.

Let's see: I use copylefed tools, and produce some program. If the
sofware is a "derivative work" of the copylefted tool (note: this may
include binaries, but not the original sources), then I can't make it
public domain, because that would involved doing the same to the parts
of the copylefted tool that it was derived from. So let's say I just
want to distributed it under the terms of the GPL. Then I have three
options:

1) Give people all the sources needed to build that tool, all
copylefted. I have no further obligations to them.

2) Give them only binary, along with a written offer to provide
sources for up to three years. If you take this route, that could put
someone in the bind you mention.

3) accompany it with the information you received as to where the
corresponding source code may be obtained.  (This alternative is
allowed only for noncommercial distribution and only if you
received the program in object code or executable form alone.)

In other words, someone is only stuck with being in a position of
having to provide source to the copylefted tools after they choose not
to if they chose the second option. The first option works quite well
for many of these people. Normally, your sources isn't a derivative
work (even if your binary isn't), so you don't have to provide sources
to the copylefted tools - just sources to your program.

So where's the problem?

	<mike
--
ICUROK2C, ICUROK2.				Mike Meyer
ICUROK2C, ICWR2.				mwm@berkeley.edu
URAQT, I WANT U2.				ucbvax!mwm
OO2EZ, I WANT U2.				mwm@ucbjade.BITNET