🚀 go-pugleaf

nelson@sun.soe.clarkson.edu (Russ Nelson) writes:
>In article <1990Mar27.193010.22308@brutus.cs.uiuc.edu> coolidge@casca.cs.uiuc.edu (John Coolidge) writes:

>   FSF asserts that any program linked with libg++ is covered by the
>   copyleft. Morally and legally speaking, this is almost certainly wrong:
>   libg++ continues to be covered by the copyleft, and must be distributed
>   with source, but the other code comprising the package remains the
>   author's property, and they can decide whether or not to offer source
>   code. Forcing a waiver of property rights, as the copyleft sometimes
>   tries to do, is about as ethically upright as if the car companies were
>   to say "any money you make from using our car (sales gained through
>   travel, salary from a job you drove to, etc) belongs to us. You choose
>   to drive the car...".

>One way to look at the copyleft is that it requires royalties equal to
>any profit you may have gained if you could have restricted
>redistribution of their code.  Apparently you think this is too high a
>royalty to pay.  That does not make it "morally and legally wrong".

The first obvious problem with this view is, of course, that I have
_in no way_ restricted redistribution of _their code_. I have merely
restricted redistribution of _my code_, which happens to be linked
to their code. There is a major difference. What people seem to be
forgetting here is that it's _my code_ I'm concerned with, not the
FSF's. If I hand someone some object files, and they link with FSF
libraries, I can't be required to give out source code. If I hand
someone a binary file already linked with FSF code, my argument is
that the situation is exactly the same: they can require source to
the FSF code, which is covered by the copyleft, but they _cannot_
require source to my code, because it is not covered by the copyleft.

My contention is that the state of affairs (both ethically and legally)
is governed by the following clause in the copyleft:

>Mere aggregation of another independent work with the Program (or its
>derivative) on a volume of a storage or distribution medium does not bring
>the other work under the scope of these terms.

Linking together my program "another independent work" with a piece
of FSF code, IMHO, constitutes "mere aggregation". This is obviously
not the _intent_ of the copyleft, but it is (IMHO, again) both
ethically and legally the correct interpretation of the copyleft.

The difference between GNUware and proprietary libraries which charge
for use is simple: the library is not freely redistributable. I _can't_
give it to someone else without paying for it. That's what library
use fees are: payments for me giving the library to someone else.
On the other hand, GNUware _is_ redistributable for free. Thus, I can
hand someone the library, since the FSF's copyright allows my to copy
the program. On the other hand, in either case (commercial library or
FSFware code) I still hold my copyright to my code. I don't have to give
it to someone else. The copyleft (or the vendor's copyright) doesn't have
the legal force to make me give it to someone else, because it's just
their copyright on their code. Copyright law _cannot_ legally be used
to force someone to abrogate their rights under copyright. Licenses
can be used to this purpose, but there's no legal license between me
and the FSF, and no vendor I've ever heard of has been willing to
put such a clause (affecting author's copyright) into a license. It
might or might not hold up; in any case, no author would sign.

Thus, I reiterate my position. The idea that a program linked with
libg++ (or any piece of FSF code) is bound by the copyleft is legally 
wrongheaded. Only the FSF code is covered by the copyleft. Furthurmore,
such a binding, if it _were_ to be legally enforcable, would constitute
an unethical intrusion into the rights under copyright of those it
affects.

--John

--------------------------------------------------------------------------
John L. Coolidge     Internet:coolidge@cs.uiuc.edu   UUCP:uiucdcs!coolidge
Of course I don't speak for the U of I (or anyone else except myself)
Copyright 1990 John L. Coolidge. Copying allowed if (and only if) attributed.
You may redistribute this article if and only if your recipients may as well.