Re: [gnugo-devel] GNU Go 2.6 compilation issue on Debian GNU/Linux

[Daniel Bump]

> Yes, I am certainly not suggesting it's a good use of _anyone's_ time
> to make it weaker :) But I have enjoyed its style of play greatly and
> would love to help in any way I can.  Is there any running list of
> bugs or features that current development would entail?  Like a view
> similar to Issues on GitHub?  I'm new to the GNU community.

There used to be a task list, and the texinfo doc mentions it.

However I can't find the task list for some reason, and
it is likely to be obsolete.

During the development years a lot of attention was given to hacking the
contents of the patterns directory, which suggest moves for GNU Go to
try. Some of the players contributing patterns were quite strong.

In a modern Go program, instead of getting moves this
way, one would use a neural net. Since processors are
faster, a lot more moves would be tried. If a version
of GNU Go using one or more neural nets is developed,
that could be interesting.

The other thing would be to revisit Gunnar's idea about 9x9 Monte Carlo
methods. It might be possible to make a GNU Go that is very strong on
9x9 games since processors are stronger nowadays.

I would suggest familiarizing yourself with the code
by reading the texinfo doc and if we can answer questions
feel free to ask.

Dan