Re: Licence questions

[Detlef Vollmann]

Paul Eggert wrote:
> 
> Detlef Vollmann <address@hidden> writes:
> 
> > GNU software that produces other software don't consider this
> > produced software as "derived" work.
> 
> That's true for some cases, but not for all.
Bison being the (only?) exception :-(

> > I'm not going to ask for an exception.
> 
> This does not sound promising.  (You want to ask us for an exception,
> but not Boost? :-)
Actually, I don't ask you for an exception.  I want you to
come back to the general rule that produced source code
is not restricted.

> > If you insist to restrict software produced by bison through
> > lalr1.cc and skeletons derived from that, I'll derive my
> > skeleton from yacc.c.  And if you think that then the produced
> > software is still restricted by GPL,
> 
> As long as Bison is generating C code for LALR(1) parsers, you fall
> within the special conditions for using Bison that exempt you from the
> GPL.  Please see <http://www.gnu.org/manual/bison/html_node/Conditions.html>
> for details.  If you are generating C++ code that is not also C code,
> then the special conditions do not apply, and the output can be
> redistributed only under the GPL.
Are you saying that if I provide my own skeleton from scratch
and only use the tables generated by bison, the source code
produced by my own skeleton is still restricted by your copyright?
That is, as if you claim a copyright on the result produced by
a calculator.  Or as if Microsoft would claim copyright to
tables produced by Excel.
IANAL, but probably that is at least in Europe even legally nonsense.

Detlef

PS: I have only heard from you.  Is that the "official" opinion
of the bison development team or just your personal opinion?

-- 
Detlef Vollmann   vollmann engineering gmbh
Linux and C++ for Embedded Systems    http://www.vollmann.ch/