The intent of my question is not for some arbitrary v
that breaks the equivalence, but for a v with some
practical use that breaks the equivalence.

The thing with $ $ v@, is, here you have an argument
with structure.  v@, discards the structure before
applying v , but after that is restructured back
to the same shape.  (Is the structure necessary
or not?)  It feels peculiar.



----- Original Message -----
From: Viktor Cerovski <[email protected]>
Date: Monday, October 4, 2010 8:41
Subject: Re: [Jprogramming] Another Abbreviation
To: [email protected]

> 
> 
> Roger Hui wrote:
> > 
> > I suspect v"0 would serve.  If it does not,
> > I'd like to see what such a v is.
> > 
> > 
> > 
> > ----- Original Message -----
> > From: Justin Paston-Cooper <[email protected]>
> > Date: Friday, October 1, 2010 9:12
> > Subject: [Jprogramming] Another Abbreviation
> > To: Programming forum <[email protected]>
> > 
> >> Hello,
> >> 
> >> Is there an abbreviation or better way of doing <<($ $ 
> >> v@,)>>, where v
> >> is a monad?
> > ---------------------------------------------------------------
> -------
> > For information about J forums see 
> http://www.jsoftware.com/forums.htm> 
> > 
> I think this much is clear: If v returns scalar, then (monadic) 
> v"0 is 
> equivalent to ($ $ v@,) and if result of v has rank > 0, then it 
> is not.
> 
> For instance:
> 
>    a =: 1 : '$ $ u@,'
>    v =: 5&,
> 
>    v"0 i. 2 2
> 5 0
> 5 1
> 
> 5 2
> 5 3
> 
>    v a i. 2 2
> 5 0
> 1 2
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to