On Mon, Mar 2, 2009 at 8:30 AM, June Kim <[email protected]> wrote: > On Tue, Mar 3, 2009 at 12:16 AM, Raul Miller <[email protected]> wrote: >> On Sun, Mar 1, 2009 at 11:29 PM, June Kim <[email protected]> wrote: >>> Tracy Harms recently put his version of cyclef at >>> http://twitter.com/kaleidic/status/1265468079 in response to Jason >> >> Does cyclef solve some useful problem? >> >> It seems to me that >> u cyclef^:# >> should, in most useful cases be equivalent to >> u > > Yes. I admit that cyclef wouldn't be very useful in J. I did it for > improving my tacit-fying ability -- I couldn't see an obvious tacit > implementation for that adverb, so I tried it. > > Many functional languages, like Lisp, Haskell and Erlang, take > lists(one-way traversal), not arrays(random-access), as their > fundamental data structure. cyclef would be a useful asset under this > circumstance. >
Thank you, June, for clarifying this difference between lists, in such languages, against what I'm used to with arrays. J has taught me to think of all rectangular arrays as lists, but now I see that this (top-of-shape) "list" aspect still has an all-at-once quality that contrasts with what others mean when they talk about lists. Maybe this will get me closer to understanding the non-determinacy model that's part of the Haskell vision of lists. Or, maybe not. Either way, I've been increasingly appreciating the ways in which J is simple. So often I find myself reading things that strive to accomplish something that is "given" in J, something you not only don't need to code, but don't need to think about (once your assumptions involve the inherent patterns). What Raul articulated is another example of this. Tracy ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
