Re: Draft new DFSG - r1.4

[john@dhh.gt.org]

Jules writes:
> To the extent that your research is a 'derived' work of SWI-Prolog, this
> restriction is perfectly admissible.

To the extent that your research is not a 'derived' work of SWI-Prolog, it
isn't.  I read this as saying that if I publish a paper on AI which
includes some prolog code that I tested using SWI-Prolog I must credit them
even though my publication does not contain one line of SWI-Prolog.

> Also, it is clearly routine academic behaviour to do so.  A piece of
> research is worthless unless it explains how to replicate the results,
> and part of this explanation will necessarily be 'appropriate
> acknowledgements and references'.

Correct.  So why demand that people do something they will do anyway?

> So, in some sense this is a 'non-restriction'.

It is a restriction on a field of endeavor.

> Maybe we should exempt this (it doesn't impinge on the freeness of the
> software, as a piece of software).

It places a requirement on those who "base research on SWI-Prolog and
publish on this research" that it does not place on others.  I suspect that
the authors really meant that they want credit when portions of their work
are included in the publication.  Why not try to convince them to clarify
the license?  Pehaps they would be willing to change the demand to a
request, or to make it clear that they mean derived works.

I don't see this as an urgent problem that would justify removing that
package.
-- 
John Hasler
john@dhh.gt.org (John Hasler)
Dancing Horse Hill
Elmwood, WI