2012/2/26 Jose Mario Quintana <josemarioquint...@2bestsystems.com>:
> That conjunction is useful but as an alternative to (`) instead of a
> replacement otherwise the definition of the also useful (`'') would be
> longer, for example, ( (`*)(((("_)(,@:{.@:))(` _))(`:6)) ) where (`) is its
> replacement (although an improved shorter definition might be possible it
> would still be longer than (`'') ).
>
> I believe Thomas is correct in his analysis (unless a higher authority states
> otherwise). I can understand now that I can manually produce a lean linear
> representation (a lot shorter than the general workaround method) of a
> trouble-making adverb that I have written because I know what I meant but
> (5!:5) does not (if you know what I mean). However, I am still somewhat
> confused: Why the atomic representation (5!:1) and (`:6) or (5!:0) have no
> difficulties? At any rate, if I recall correctly, it has been stated that
> (13 :) generates tacit equivalents of explicit verbs based on how they would
> be executed. I wonder if a similar approach could be used to generate a lean
> linear representation of a trouble-making adverb.
I am sure that 5!:0 has support code for each primitive. Thus, to
convert from atomic representation to linear representation
name=. atomicRep 5!:0
5!:5 <'name'
Or you could build your own lookup table with every primitive represented.
--
Raul
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
