On Wed, 6 Jun 2001, "Thomas Charron" <[EMAIL PROTECTED]> wrote:
>
> The only thing that really worried me, after rereading it myself, is
> this:
>
> <SNIP>
> #37: "If modules are designed to run linked together in a shared address
> space, that almost surely means combining them into one program.
> </SNIP>
>
> *sigh*. Thats from the FAQ. Blech. Whoever added that was very keen
> at expressing their opinion.
I've been thinking about this aspect of the GPL recently. I feel this
vagueness & ambiguity about the definition of "linking" is a sleeping
problem for it...
I know that Stallman and FSF say otherwise, but I even feel if I write a
proprietary program that calls, say, gzip (via system or fork+exec)
then my proprietary program is "a work based" on the gpl'd gzip. And
so I can't distribute my program with proprietary-type license.
(gzip is not a great example, I know about libz.so, you get the idea:
proprietary program runs GPL'd program).
I am worried that the GPL does not spell out this fork+exec is explicitly
permitted, this is a crack that lawyers can insert a wedge and start
pounding...
Any thoughts on this?
Even more bizarre, if we suppose that a proprietary program calling GPL
program is OK, what if I write a GPL'd program that links in GPL'd
libraries (not LGPL) and (awkwardly and inefficiently, but
nevertheless) somehow provides the "services" of the GPL'd library to
the proprietary program?
Karl
**********************************************************
To unsubscribe from this list, send mail to
[EMAIL PROTECTED] with the following text in the
*body* (*not* the subject line) of the letter:
unsubscribe gnhlug
**********************************************************
