Must do:

- WIN32: define a type for __int64 and/or long long int, possibly
  using <stdint.h> (POSIX compliant, does windows have this?) to make
  a portable solution for the long long int code. Grrrr, Micro$oft!

  -> Done, except for testing if the MSVC detection code works, and I
     am not sure how to do an unsigned __int64 in MSVC.

- Proper installation in Makefile with installation in system
  directories, if desired.

- Windows project files updated, + installation instructions
  (everything in 1 dir is easiest here)

- Binaries compiled, everything uploaded to allegro.cc and my
  website. Make a small website with screenshots.

- Portability tests in general. Build and run on at least
  windows/msvc/mingw, linux, DOS/DJGPP. Find out if the mkdir code in
  file.cpp is portable on platforms like borland.

Wishlist:

- a high score list, or at least a list of old finished games. This
  requires input of player names if I want to do it properly, and
  therefore another dialog.

- sound, if I can come up with a good sound for placing stones and
  flipping stones. I don't really think so, but who knows. Spoken text
  (Game Over, etc) might be fun.

  -> partially done, some obvious error messages were removed and
     replaced by a simple 'beep'.

- a gameclock, and the option to set a limited amount of thinking time
  for each player. Go through your time -> you loose.

- Some decent matches against multiple othello programs would be
  nice. Kreversi wins at ply 4, but looses at ply 6. I still need to
  try out what happens at the highest ply level. Ryxx (?) othello
  lost, according to Richard. Gnome othello is no challenge, even I
  can win from that at it's most difficult setting.  Xreversi seems to
  be a decent program (minimax algorithm, corner and edge patterns)
  but it either has a bug or a faulty scoring algorithm, because my
  program (and myself, sort of) manages to win in the endgame,
  probably because xreversi doesn't switch to counting stones near the
  end of the game.

- Mention the program to the author of the (french) reversi website,
  who keeps a list of reversi games. I hope to get him to include this
  one, considering iagno and kreversi are on there that should be
  possible.

- Learn to beat my own program at ply 8!

Very far future:

- pattern recognition and opening books would be cool to make to
  improve the playing strength of this game, but considering it
  already beats the *&(^*& out of me at ply 4, it is pretty hard for
  me to estimate when the program is improving or becoming worse in
  playing strength. This will require some more knowledge and examples
  about othello, how to play it and how to write computer algorithms.
