Re: Bug#227159: ocaml: license conflict in Emacs Lisp support?
On Mon, Jan 12, 2004 at 11:10:00PM -0500, Walter Landry wrote:
> Sven Luther <firstname.lastname@example.org> wrote:
> > On Mon, Jan 12, 2004 at 08:53:42AM -0500, Walter Landry wrote:
> > > DFSG #2:
> > >
> > > The program must include source code, and must allow distribution in
> > > source code as well as compiled form.
> > But we don't do distribute compiled forms, and it doesn't really make
> > sense to do so.
> Note that DFSG #2 says "must allow distribution". Debian is not
> allowed to distribute binaries, regardless of whether it would like
Yep, Debian is. There is absolutely no reason to distribute compiled
.elc files, even if we don't do it. The only restriction is on linking,
which will only happen on the users machine anyway.
> > > It sounds like we can't distribute compiled forms. You could put this
> > > in non-free, since we can distribute the source. In that case, I am
> > Ah, but we are going to remove non-free anyway, so i clearly won't do
> > that.
> > > uncertain about whether you should disable the automatic generation of
> > > .elc files.
> > Why ? We clearly are not violating the GPL by doing so, so where is the
> > problem.
> If Debian sets up everything so that the user automatically makes the
> link in the postinst, a judge might see that as legally equivalent to
Nope, not in the postinst, but only when launching emacs.
> distributing the compiled form. Especially since Debian distributes
> Emacs as well. It gets rather murky, and starts getting into what
> people's intent is. It might be fine, but I am uncertain.
Mmm, i don't know legal stuff so well, so i cannot say. My impression is
that a judge will laugh in your face if we ever come to him about such a
petty complaint, but well.
> > > So I think talking to the upstream is a good idea.
> > Sure, but on more serious ground than this. Notice that the bugreport
> > claims that RMS thinks that ..., not that it is actually true.
> Regardless of what RMS thinks, do you think that the compiled forms
> are legally distributable? I am far from an expert on lisp compilers,
Yes. They are just compiled bytecode, not linked with emacs in any way.
If i understood it well, they have even less emacs in them than .o files
have gcc code in them, and gcc has a special excemption for libcrt.o or
whatever this stuff is named that goes into all binary files produced by
> but I would think that the lisp compiler mingled itself with the code
> just as much as a C compiler.
I fear this is not the case. And anyway, this is then the lisp compiler,
not emacs we are then speaking about.
Anyway, i will write an email to upstream about this all by myself,
thanks for your time though.