Hi folks,
(a bit daring to send it to so many people/lists, but I guess that one reason why not so much technical related etoys stuff is sent, is because people are too shy to send their comments to more than one list, so I hope this is ok...;-))
After some experiments I found the trick to do a Tic Tac Toe-game entirely with Etoys. At least a version with human only players, but I nevertheless want it to share with you:
http://www.squeakland.org/project.jsp?http://www.squeakalpha.org:8080/ super/uploads/TicTacToe.002.pr
It is altogether 43 lines of code, including the script-headers.
The idea is to draw some morphs (polygons) which represent the winning lines. When a stone is set, the polygons are asked if they overlap it. If yes, they add the value of this stone to their internal counter, if this counter is not 0 and divisible by 3 we have a winning situation. The cross is worth +1 and the circle worth -1. Also the number of already set stones is counted so that a draw can be detected. Enhancing this to 4x4 is quite easy, at least if 4 in a row shall win then.
But if someone has any idea for Gomoku without ending up by having to draw a zillion of lines, I'd be interested. One idea in that direction would be to give players some kind of "radar", where they could look in some specific direction so that they can see all their neighbors (of some kind?) in a distance of 1 to n cells, assumed they live in a matrix of cells. As directions there would be only vertical, horizontal and diagonal. Simulations like the game of life and general cellular automatons should become very easy to do with this approach, so anybody up to give our deaf, dumb and blind players some fast sensors? I did a first prototype for that, which is darn slow, but I'd be happy to share that code for making it fast. Or are there other tricks/views concerning this problem on the way?
Have fun,
Markus Gaelli
For a funny tic tac toe with an analogue computer opponent see: http://www.rci.rutgers.edu/~cfs/472_html/Intro/TinkertoyComputer/ TinkerToy.html