[ List Earliest Comments Only For Pages | Games | Rated Pages | Rated Games | Subjects of Discussion ]
Check out Janggi (Korean Chess), our featured variant for December, 2024.
Check out Janggi (Korean Chess), our featured variant for December, 2024.
I wonder whether it would do any harm to redefine the meaning of y as just indicating a range toggle, rather than m with a range toggle. The meaning of the combination ya would remain the same, as a without any explicit mode indicator defaults to ma. Only in combination with an explicit mode it would make a difference: currently ycaX means yaXcaX, i.e. a range toggle on an empty square, or a locust capture without range toggle. While the new meaning would be a locust capture with range toggle, without the possibility to end the first leg on an empty square.
The point is that I cannot imagine a combination of a range-toggling mode and a different non-toggling mode would ever be needed. In fact it appears that in the XBetza implementation of the Interactive Diagram (and its conversion to GAME code) such combinations don't even work as the old definition of y required. Due to a bug the range toggling specified by y already 'leaks' into modes that were specified as alternatives to it, and up to recently this was never noticed.
This redefinition of y removes the need to resort to using non-alphabetic characters for now, as ya takes the function of the proposed a'. The modifier t can then be kept in the proposed meaning of 'terminate', (as alternative to a) to indicate the remaining legs of the path are not performed by the piece, but just administers its 'influence' (such as burning). On second thought using an explicit indicator for the start of the 'influence path' is more versatile than using a separate mode (like cc) for burning, because it also allows burning according to a multi-leg path. The latter might be needed to emulate burning of non-adjacent pieces through a (dummy or lame) intermediate square.
With this new meaning of y, the Advancer would be ytfcQ. An Advancer that stops 2 squares before its victim would be ytfafcQ, and one that captures the first piece in the direction it moves tfcQ.
I also gave some thought to the issue of 'initial' moves. It is reasonably common that such moves depend not on whether the piece was moved before, but on whether it is in its initial location. E.g. for Pawns in Crazyhouse, which have the double push on the 2nd rank even when they were moved, captured and dropped before. This interpretation of 'initial' could be indicated by a leading ii on the move, instead of just a single i.