Ok, with all the various opinions, I think I'll:
o Stick with the State monad.
o Switch from |> to $ and teach readers how to read it, "Think of 'f $
g $ x' as 'f of g of x' or 'f(g(x))'. From that point of view, it may
be helpful to read 'f $ g $ x' from right to left."
Unless there are any objections, with that one change, I'll consider
the coding done and move on to writing the article.
Thanks so much for all of your various opinions and suggestions! I
feel much more comfortable speaking from a position of authority
knowing that all of you have reviewed my code!
Best Regards,
-jj
On 3/15/06, Udo Stenzel
Shannon -jj Behrens wrote:
o How important is it that I switch from using the State monad to using arrows?
Your problem seems to be naturally soved by the State monad, therefore you should use that.
o How important is it that I switch from using |> or $ to using arrows?
Unimportant. However, I'd recommend switching from application ($,|>) to composition (.,>>>) where possible. It's "more functional" and often easier to read.
o How much will this increase the "conceptual complexity" of my program
Not at all. You might define >>> locally as
f >>> g = \x -> g (f x)
or just pretend that this definition is contained in Control.Arrow due to a historical accident, thereby completely ignoring the existence of other arrows.
Udo. -- Wo die Macht geistlos ist, ist der Geist machtlos. (aus einem Gipfelbuch)
-----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.1 (GNU/Linux)
iD8DBQFEF+f5c1ZCC9bsOpURAv2gAJwNirkt2yMFLlbTT9I2twUs3UcxdQCeKqx2 0FVTzx7VJEGtJexlGIJxero= =CPSW -----END PGP SIGNATURE-----