🚀 go-pugleaf

Quotation from Per Abrahamsen <abraham@iesd.auc.dk> article
<ABRAHAM.92Jan25150051@loke.iesd.auc.dk>

Per> However, if you compile a binary `a.out' from the two source
Per> files `gnu.c' and `mathew.c', then a.out will clearly be a
Per> derivative of both files.  Thus, if you distribute a.out (a
Per> derivative of `gnu.c') then you must also be willing to
Per> distribute the full source for `a.out', i.e. both `gnu.c' and
Per> `mathew.c'.  Given FSF's goal, that seems very reasonable.  But
Per> that wasn't what you asked about.

I have not read the entire GPL today, but if a.out is considered a
derivate of gnu.c, then surely it must suffice to provide a.out
together with gnu.c and matthew.o. I don't see the rationale for GPL
forcing the distribution of matthew.c.

What if mathew.c is a derivate of mathew.y, does this guy then have to
deliver that mathew.y or would mathew.c suffice? The GPL does define
the term source code (in section 3), but who can prove that there was
a mathew.y before mathew.c? And who can prove that there was a
mathew.c before mathew.o?

What if Mathew sells mathew.o (without providing mathew.c), and Nancy
distributes gnu.c under GPL, and Olivia works as a consultant offering
to compile and link files for John Doe (who doesn't want to compile
them himself)? Would Mathew violate the GPL? What if Mathew and Nancy
and Olivia is the same person or works for the same company?

Obviously, M.N.O. Inc is the cause of the existence of a.out in John
Doe's computer, but is this defined as distributing a.out? I am not an
expert on law, especially not USAmerican law, so it would be
interesting to hear other opinions.

In my humble opinion, the so called Free Software Foundation should be
renamed the GPL Software Foundation. Or else, we will soon be writing
"Freedom(TM)".
--
Lars Aronsson, Lysator, Linkoping University, Linkoping, Sweden