Re: [pbmserv-dev] Re: [pbmserv] Palago notation?

["David J Bush" <twixt@cstone.net>]

| There's no randomness in the Palago space labelling scheme. For a 
| given turn, the M move positions are labelled in anticlockwise order 
| by angle relative to the tile group's centre of mass, starting with 
| the upper leftmost tile as move 1. The idea was to keep similar move 
| numbers close together. This isn't an ideal scheme as the numbers 
| need to be recomputed each turn - not much use if you want to replay 
| a game with actual tiles from a written move list, for example.

Well, sure. What purpose does notation serve, other than allowing
replay, either with a physical set or with any software which is not
part of Gamerz? Center of mass? Was this easy to program? It would
definitely not be easy to determine independently at later stages of
the game. I rate the current Palago notation as confusing at best.

| Another option would be to use the actual [row,col] grid coordinates 
| of each tile, but this would make the notation very clumsy and get 
| confusing due to the hexagonal nature of the grid. It's much more 
| elegant to specify moves as say "9b" than "[-12,17]b".

I don't know about games other than Palago, but sure, any sort of
Cartesian coordinates would result in awkward looking notation.

| Yet another option would be to use a fixed cell labelling scheme such 
| as the spiral honeycomb mosaic (SHM): 
| http://ozviz.wasp.uwa.edu.au/~pbourke/texture_colour/shm/
| I've devised a similar spiral labelling scheme for recent 
| implementations of hexagonal tiling games which seems to work well, a 
| bit like SHM but without the discontinuities:
| 
|       8   9   10
| 
|     7   1   2   11
| 
|   18  6   0   3   12
| 
|     17  5   4   13
| 
|       16  15  14
| 
| The benefit is that each number refers to exactly one grid position 
| for the entire game (and indeed for all games) rather than the move 
| position numbers appearing to jump around each turn. I'd probably use 
| this scheme in future.

Okay that looks fine with me. Of course you needn't show the entire
outermost hexagonal ring, just the vacant cells which are pertinent to
the game. I hope you will consider switching the Palago numbering scheme 
to this (rotated clockwise 30 degrees.) It's more work for you as well as
Paul van Wamelen, but without it you might as well delete the notation.

Great game BTW :-)