Re: [gnugo-devel] endgame module for GNU Go

[Evan Daniel]

I think you're misunderstanding something fairly fundamental about the
GNU Go project.  We are interested in shipping a Free, strong, and
easy to use Go playing engine.  For a definition of Free Software, see
http://www.gnu.org/philosophy/free-sw.html.  That the engine is Free
is a primary requirement, and something that the GNU Go developers see
as such, not merely an added benefit or feature tacked on the side.

This means that in order for us to be interested in including an
endgame planner, or any other piece of software, that software must
also be Free.  If you were to write an endgame module that conformed
to some sort of AI standard for a planner (which it sounds like at
least mostly exists?), and provided a Free planner that could be
plugged in, we would be interested in such provided that it was a
useful addition to the program in terms of functionality (ie made the
program stronger enough to be worth the added execution time and added
maintenence requirements).  If that planner was easily replaceable by
proprietary planners, or other Free planners, then that would be
perfectly fine, and I think there are people who would find that
useful.  However, we need to be able to ship a single Free package
that contains all pieces necessary to use the program as it is
currently used, ideally without any new external dependencies.

In order for your work to be incorporated into the main GNU Go tree,
all the above requirements would need to be met (and a few others,
probably, though I don't think anything major beyond the above).

If, however, your interest is solely to demonstrate that an AI planner
is effective at playing Go endgames, and you don't need to have the
code incorporated into the shipping version of GNU Go, then the
requirements get somewhat more relaxed.  You could, for example,
create an endgame module that linked to an AI planner for which there
was no Free planner available (provided you linked everything in a
manner compatible with the GPL, which is not something I'm an expert
on).

I think that if you are interested and able to write a piece of
software that meets the above requirements, you will find ample help
in the details of how to hook it into GNU Go and how to play Go
endgames; if you want, I can offer some comments on the technical side
of what you've mentioned so far.

Evan Daniel

On Sun, 5 Sep 2004 13:38:37 -0700 (PDT), Eric <address@hidden> wrote:
> --- Paul Pogonyshev <address@hidden> wrote:
> 
> > Eric wrote:
> >
> > > > Are you interested in getting it into
> > > > the mainline GNU Go
> > > > or do you regard it as a private research
> > project?
> > >
> > > Not really sure what you mean by this, but I'll
> > take a
> > > stab at it:
> > >
> > > The Semsyn engine itself has already been
> > completed,
> > > and is proprietary. However, most planning systems
> > > being developed these days adhere to the
> > "standard"
> > > PDDL (Planning Domain Definition Language) which
> > > originated from Yale University.
> > >
> > > If GNU Go treats the planner as a black box, then
> > it
> > > is not necessary to distribute the planner as
> > integral
> > > to the GNU Go engine. Users of GNU Go can simply
> > > "plug-in" their favorite flavor of planner.
> >
> > This does not sound acceptable.  While GNU GPL
> > allows
> > communicating with a proprietary program (i.e. when
> > that
> > program is a separate process), there is a number of
> > serious technical and other problems:
> >
> > * We cannot distribute your program (it is
> > proprietary)
> >   along with GNU Go.
> 
> There are lots of free planners available. Try to
> Google 'AI Planning Systems', since you obviously have
> not done this.
> 
> > * We cannot even put links to your program, because
> > GNU
> >   servers don't advertise proprietary software.
> >
> > * I don't think it is worth the inconvenience of
> > plugging
> >   in a planner into GNU Go.
> 
> Yea, unless it makes GNU Go useful for something
> besides your own personal entertainment.
> 
> > Most users would not
> > even
> >   know what a planner was and that GNU Go had a use
> > for
> >   one.
> 
> Duh. That's the whole purpose of the effort. Nobody
> knew what personal computers were useful for either,
> until Bill Gates starting building them.
> 
> > We hardly want to make GNU Go depend on
> > another
> >   program (even if optionally) other than a GUI.
> 
> Why not?
> 
> > > The bulk of the work being proposed currently
> > involves
> > > writing the endgame module and modelling the Go
> > > domain, i.e. designing the domain operators. And
> > yes,
> > > I intend to leave these pieces in the public
> > domain
> > > (or free, or whatever is the proper terminology).
> >
> > As far as I understood, this endgame module would
> > essentially be an interface to a planner.
> 
> Very good.
> 
> > Then, as I
> > explained above, without a free software planner
> > which
> > can be compiled into GNU Go, this is hardly of any
> > use.
> 
> I already addressed this above.
> 
> > In general, your program's being proprietory is the
> > show stopper.
> 
> No it's not.
> 
> > I understand that you will not change
> > your license because of this, but we cannot change
> > our
> > policy either.
> 
> That's your loss.
> 
> 
> 
> 
> _______________________________________________
> gnugo-devel mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/gnugo-devel
> 
>