I guess I disagree. The program should not be automatically making choices for the user. (One could include a preference setting that says "Allow gnubg to interpret ambiguous checker moves" or the like, but that would be adding yet another optional feature.) Probably the "purest" solution would be to simply have gnubg reject the move, while popping up a message such as "Ambiguous checker move." This would force the user to make a valid play, without providing information as to what the possible moves are. Another option would be to pop-up a list of valid interpretations and have the user select the one he wants.
If nothing else, would someone explain the current process that gnubg uses to interpret ambiguous checker slides? Does it choose the play that is "best" for the user? Does it assume that hitting a blot is always intended? Does it always play the smaller number first, and then the larger? The fact that a user has to ask such questions seems problematic to me. (As a last resort I could analyze the code, but for me that would be an effort!)
----- Original Message -----
From: "Jonathan Kinsey" <address@hidden>
To: address@hidden
Cc: address@hidden
Sent: Thursday, April 9, 2009 4:22:41 AM GMT -05:00 US/Canada Eastern
Subject: Re: [Bug-gnubg] Handling ambiguous checker moves
I just explicitly enter ambiguous moves, either by dragging twice or using the
left/right mouse buttons. I can't see any problem with gnubg interpreting an
ambiguous drag as a pick-n-pass as that's generally the most likely move in
these cases.
All in my opinion of course,
Jon
Zulli, Louis P wrote:
> In the position
>
> ttsFEAJvzR0AAA:QQnlAAAAAAAA
>
> I roll 2-1 and drag a checker from my 6-point to my 3-point. gnubg
> /automatically/ interprets this as 6/5*/3 rather than 6/4/3. Why? Seems
> to me that when the user makes an ambiguous move it should be flagged as
> such, and not accepted. Perhaps a list of possible interpretations
> (moves) could pop-up, from which the user can select the move he
> intends. But the program should not make a choice for the user.
>
> Is this a new issue?
>
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Bug-gnubg mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
Get the New Internet Explore 8 Optimised for MSN.
Download Now