> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
> Behalf Of Jose Mario Quintana
> Sent: Saturday, July 08, 2006 3:18 PM
> To: 'Beta forum'
> Subject: RE: [Jbeta] Incorrect result shape from 5 +"1 (0 1 $' ')
> 


> 
>    X=. 5 [ Y=. 0 1$' '
>    plus=. +
>    plus1=. plus"1
> 
>    X (plus1 f. -: plus1) Y
> 0
> 

Until recently I naively assumed that the above should not happen.  It seems
to me that the DoJ text for the 'f.' entry,

"If x is a proverb, then y=: x f. is equivalent to it, except that any names
that occur in the definition of x are (recursively) replaced by their
referents"

should be amended.  Perhaps it should be clarified that the equivalence only
holds for cases of "practical interest" although, to be meaningful, the
concept of "practical interest" should be clarified as well.

I am interested in this issue because I typically use proverbs to define
other proverbs in the developing and testing phase and tacit
(function-level) production code is ultimately generated by a fixing
process.  At this point I am not longer sure what the domain of the
equivalence is supposed to be; any clarifications (from anyone, of course)
would be much appreciated.

Going forward I could fix code as soon as possible for testing purposes but
this would complicate a lot the debugging phase as well.  Either way, I am
becoming very concerned about what I might have missed in the past.  Should
I be worried?




----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to