Raul wrote: > "it wouldn't be a natural fit" for the current implementation.
I think there’s a philosophical conflict between the two framing metaphors (arrays vs swarms/CAs). My take is “J swarm processing capabilities” would not be J. It would be a new language (maybe a very interesting new language, maybe not, but a new language in any case). > For example: some primitives which work on dense arrays do not > work on sparse arrays. > > For example: 13 : falls back to 3 : or 4 : for some expressions. This would not be true of a “J with full sparse array support” nor a “J with a perfect 13 : ‘’ transliterator which never falls back on 3 : or 4 : “. Those mechanisms are completely compatible with the fundamental array-orientation of J. -Dan ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
