Re: Compiled files without sources????

[David Kastrup <dak@gnu.org>]

"Stephen J. Turnbull" <stephen@xemacs.org> writes:

> Tim Cross writes:
>
>  > I think it is far more likely the error was due to an honest error
>  > in understanding/interpretation of the licensing requirements
>  > rather than the somewhat malicious and deliberate interpretation
>  > you have suggested.
>
> No malice at all is implied, just a lack of due diligence.  Obviously
> there was an error in understanding the implications of the licensing
> requirements for what Emacs *should* do (though Emacs is in no way
> legally bound, of course).

The FSF, being the principal copyright holder of Emacs, is not bound to
the conditions of the GPL in a legal sense, but setting a bad example
like that is not going to increase the respect from others.

The FSF is bound to the conditions of the copyright assignment contracts
from various contributors, however.  While they are free to pick a
number of free licensing arrangements for Emacs code not restricted to
the GPL, they are not free to distribute generated code without source.

People refusing to license under a GPLv$$+ scheme or assign their
copyright to the FSF because either would be tantamount to a blank
cheque and the FSF could turn bad, tend to forget that the FSF has
entered into legally binding agreements as well.  Even if the FSF became
a subsidiary of the Gates foundation tomorrow, nothing really bad could
happen to the software for which the FSF signed contracts.

So "Emacs" (which is not a sentient entity yet) is certainly not just
morally, but also legally bound to providing corresponding source code
to its binaries.

> Deliberate, yes.  In http://article.gmane.org/gmane.emacs.devel/115057
> Chong Yidong writes: "I think it's better for us to merge just the
> generated Lisp grammar files, leaving the grammar development for
> upstream.  It's an awful lot of infrastructure to pull in...."  IMO,
> "just the generated files" should be a red flag for any maintainer
> working with copyleft licensing.

Well, you have the advantage of long years of training looking for such
pitfalls as project/release manager of XEmacs.  Emacs has had less
reason or opportunity to try staying under the radar and/or out of reach
of Richard.  Perhaps you should offer courses.

-- 
David Kastrup