On Tue, 2004-09-07 at 12:21, Jacques GARRIGUE wrote:

> 2) the point I was trying to make was that the distinction between
>    static and dynamic linkage is not relevant here, since the tool 
>    discussed does not link to anything anyway.

As it stands. But a client might want to modify it.
They might want to bind more strongly to it.
They might want to add a custom routine to it.

See below for a real example.

> An ambiguity is only a problem if you happen to be in the ambiguous
> area. I do not see how it can matter when you are well outside of it.

Because things change -- what starts off clear may become
less clear as you modify things.

Here is a real example. Felix contains the CIL package,
which is a C parser. It is used in a stand alone wrapper
generator tool. Suppose CIL were GPL***, and I don't care
if the wrapper generator is GPL. All is fine!

Only -- a cute trick is to actually build the wrapper
generator directly into the Felix compiler so you can
#include a C file directly (and the compiler processes
the generated wrappers).

I actually DID do this (although I removed that feature).
Well, if CIL were GPL that would make my core compiler
also GPL -- simply because of a small change in the
integration architecture.

Such absurd artificial constraints in an experimental
package make multi-licence situations undesirable.
It also isn't just a legal problem that I might be
able to sort out -- there is the issue of what
potential clients might think with a cursory
glance at the project.

*** CIL is BSD. If it had been GPL I wouldn't have
even considered it for inclusion.

-- 
John Skaller, mailto:skaller@users.sf.net
voice: 061-2-9660-0850, 
snail: PO BOX 401 Glebe NSW 2037 Australia
Checkout the Felix programming language http://felix.sf.net



-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners