Re: I'm soliciting requests for aisleriot games

[Alan Horkan <horkana maths tcd ie>]

On Sun, 18 Feb 2007, Vincent Povirk wrote:

> On 2/18/07, Alan Horkan <horkana maths tcd ie> wrote:

> > There were some game types I was interested in, such as Reverse versions
> > of existing games.  If you have been playing a game and trying to get the

I also tend to favour the easier variations and games which could be
played quickly and often rather than very difficult versions requiring
long and careful deliberation.

(In one of the bug reports I suggested a much easeir Golf variation called
Putt Putt and my request was misinterpeted as a call to make the game
easier as opposed to adding a variation.)

> > cards out in order from ace to king, to then go and play a version of
> > the game where you need to play backwards and get the cards out in
> > order from king to ace can be quite confusing and seem more difficult even
> > though it is only equally difficult.  Could take quite a while to do
> > something like that with a few of the more popular games.

> This seems like a waste of time to me,

I mention it specifically because I greatly enjoyed playing both Freecell
and Inverted Freecell.  The confusion of switching from one to the other
made things more challenging, a bit like trying to play snooker left
handed.

> since reversing the card order creates an identical game or,

> if it's done for all games,

wouldn't be suggesting it for more than a handful of games

> a confusing and not very useful new option.

> > I'll certianly give it more thought.  (Dig out my old copy of goodsol.)

> There are all sorts of lists of solitaire games I could be using, but

Pretty Good Solitaire (goodsol) has hundreds more variations than
Aisleriot but some are really very slight variations and not particularly
interesting.

Having said that games like Klondike differ very little from games like
Thumb and Pouch but there is no question of removing either of them, lots
of code could be consolidated though.

> I don't want to do that. I only want to bother with a game if I know
> someone would play it.

I'd be careful to suggest only games I enjoyed playing a lot or thought
would be quite easy yet interesting variations of existing games.

> I'm also discouraged by the inconsistent and often confusing/ambiguous
> rules documentation I find on the Internet and would rather have a human
> around to answer questions about the rules.

Understandable and I wouldn't push you to work on a game you didn't enjoy
but given how these games have incosistent rules in the real world it
might not always be possible to get clear and definative rules.  I
understood "solitaire" or patience to mean 3 card Klondike, where you
would switch to single card dealing if you got stuck but I've learned
others have quite different rules which match up to other variations they
didn't know the official names of either.  (Which reminds me of one
rather interesting variation of Klondike called Gold rush, where you start
by dealing 3 cards each time, then when the deck runs out dealing 2 at a
time, then on the last turn you deal the cards 1 by 1.)

> > Also if you were interested anything you can do that would make it easier
> > to add Autoplay to a game would be great. When I wrote my few new games I
> > did make an effort to add more abstractions for things like the tableau
> > and the foundations so that a game could more easily be reused to create
> > another game with a different deck size.  Even more abstractions and
> > cleanup would be need to allow for autoplay, and with that kind of
> > abstraction you could slim down many of the existing games and make it so
> > much easeir to write new variations.

> Hmm, we do need to make autoplay more consistent (and documented? is
> it documented anywhere? maybe it needs to be consistent before it can
> be documented). Maybe I'll work on that and/or file a bug about it.
>
> As I understand it most games use the convention where a double click
> means a move to foundation, so autoplay should be possible by
> simulating double clicks on the piles with cards that could be played
> there.

I dont think you can get away with making it quite that simple, you need
to be careful about not playing ahead too far as some variations do not
allow cards to be played back from the foundations.

> I think there's a helper function for that. The only cleanup I
> can think of that would help on most games would be to replace
> hard-coded lists of slots with names like "tableau" and "foundation".

I was tempted to do that and at least make things easier if anyone else
came along and wanted to take another step but I never got around to it.
As I recall I was also interestd in making games more generic so that
something like a 2 deck version would only require a tiny difference in
code in which case the abstractions are particularly useful.

> I'd rather not add any new abstractions until I know someone will use
> them (such as me adding a game). I'm just too terrible at predicting
> ahead of time what sort of thing will be useful.

Consolidating the existing games might be a better way to go, but only if
that really interests you.  Adding new games is a benefit users are much
more likely to see and appreciate but I'll be happy enough just to see
someone working on Aisleriot adding new games or making it easier to add
them.

Thanks Vincent.

Sincerely

Alan Horkan

http://advogato.org/person/AlanHorkan/
http://www.linkedin.com/in/alanhorkan
http://alanhorkan.livejournal.com/