[Cuis] Solitaire updated for 2203

Ken.Dickey Ken.Dickey at whidbey.com
Wed Mar 25 15:18:12 CDT 2015


On Wed, 25 Mar 2015 09:12:54 -0300
Juan Vuletich <juan at jvuletich.org> wrote:
..
> You managed to do the animation and game logic, interleaved with normal 
> World cycling, and without any need to mess with "private" Morphic 
> internals. Impressive.
> 
> I'm not sure, but I have a feeling that making 'mainLoop' and 
> 'pilesLoop' into separate methods might make debugging a bit easier.

Proof of concept code is not always elegant.  

First step.  "If you can do it, it must be possible."

> I also believe that once you leave the world of simple lineal sequencing 
> of calls and returns, debugging will always get trickier, just as design 
> and coding. At least, your code is entirely predictable and 
> deterministic.

The CPS transform/style itself is entirely deterministic.  Stepping adds some non-determinism but not much in this context.  

Disabeling the Undo button during undo helps.  ;^)

I need to look at the SeaSide code base to see what code/style they have developed to handle continuation based processing.  They have been using continuations for some time to deal with web page processing.

No time as yet.


Cheers,
-KenD




More information about the Cuis mailing list