[Retros] Singular PG

[raosorio at fibertel.com.ar]

This message was sent on Tuesday 14th but it does not appear yet.
Just in case I resend it.

Dear friends,

I had some feedback through my personal mail. Here follows,

1) Proposed problem:
I apologyze because I presented an old version (incorrect).
Thanks to Peter va den Heuvel who made me see the mistake,
The correct one is,

rnb2bnk/pppp1p2/1q1p4/8/7p/2N5/PQP1P1PP/R1B1KBNR
PG in 22 single moves Singular on GOS

2) Tries:
Andrei Frolkin was specially interested in this potential.
He wrote,

The most interesting implementation of the idea would be to have a try which is refuted FOR ONE REASON ONLY. For example, in case of an SPG in N moves, if you suggest that it might be Madrasi then there is a UNIQUE sequence which stops at move (N-1) for a truly Madrasi reason. Then it turns out that the sought-for type is..... 

3) Future stipulations:
Thierry le Glehuer wrote,
My first opinion is that you have to consider the possibility to create
a new fancy chess rule in the future. So your problem could be cooked
tomorrow dispite he has only one solution today!

My answer was,
Good point. For this reasson, the GOS list should be explicit, i.e., listing one by one the stipulations that includes. There would be no problem at the beggining, since it's just a conventional thing. After, any new stipulation should be critically analyzed in order to decide if it is included or not. In principle, there should  be no problems since a really new stipulation would have its own particualrity that is not expected to create conflicts against the existing ones (except if it is a light variation of an existing one)


4) Two phases (two solutions/twins)
It seems to be a challenge to get two phases, one stipulation each.

Best,
Roberto