Thank you.

The reason why I asked is that the dyad u/ was defined
and set in concrete before we really understood rank,
and now that we do, and knowing that u/ is simply u"(lv,_),
we would probably use u/ for something else
(/ is prime real estate, being a single character word).
That is, if we had any choice (which we don't).

I think it was Larry Breed who said that "give me the 
symbol for a function and I'll come up with a mnemonic".
So your explanation is as good as any.  (It is different
from Ken's original reasoning, I think.)  I note that the
dyad u/ did not have its current meaning in 1990.
See the table in Section A of 
http://www.jsoftware.com/papers/J1990.htm



----- Original Message -----
From: Graham Parkhouse <[email protected]>
Date: Thursday, February 25, 2010 2:22
Subject: Re: [Jprogramming] Agreement
To: [email protected]

> Why do I think "table" is [so appropriately] the dyad of u/ 
> (whose monad is
> insert)? They both represent fundamental rank changing 
> operations, one
> (contraction) whereby a dimension is absorbed and the rank is 
> reduced by
> one, the other (cartesian product) whereby dimensions are 
> concatenated and
> ranks are summed. OK, not such a blinding insight, but these 
> types of rank
> changing operation (including ,. ,:) are what makes array manipulation
> useful.
> 
> All benefits from array manipulation come *only* when there is a 
> change in
> rank.
> 
> Does this make sense?
> 
> Regards
> 
> Graham   
> 
> 
> 
> ----- Original Message -----
> 
> > Date: Wed, 24 Feb 2010 07:30:10 -0800
> > From: Roger Hui <[email protected]>
> > Subject: Re: [Jprogramming] Agreement
> > To: Programming forum <[email protected]>
> > Message-ID: <[email protected]>
> > Content-Type: text/plain; charset=iso-8859-1
> > 
> > Now you've got me curious.  Why do you think "table"
> > is the dyad of u/ (whose monad is insert)?
> > 
> > (There is no right or wrong answer.)
> > 
> > 
> > 
> > ----- Original Message -----
> > From: Graham Parkhouse <[email protected]>
> > Date: Wednesday, February 24, 2010 5:04
> > Subject: Re: [Jprogramming] Agreement
> > To: [email protected]
> > 
> > > Thank you, Roger! I'm very happy with that. I have *just*
> > > understood why
> > > table is the dyadic equivalent of insert. That's brilliant! Till
> > > now, I had
> > > just seen it as two useful verbs using the same symbols.
> > >
> > > Regards
> > >
> > > Graham
> > >
> > >
> > >
> > > ?----- Original Message -----
> > > >
> > > > Message: 6
> > > > Date: Tue, 23 Feb 2010 10:11:02 -0800
> > > > From: Roger Hui <[email protected]>
> > > > Subject: Re: [Jprogramming] Agreement
> > > > To: Programming forum <[email protected]>
> > > > Message-ID: <[email protected]>
> > > > Content-Type: text/plain; charset=iso-8859-1
> > > >
> > > > If v has ranks 0, x v"0 _ y or x v/y should have shape 
> ($x),($y),s> > > where s is the maximal shape of the individual 
> results.> > > If v always returns a rank 0 result than that simplifies
> > > > to ($x),$y .
> > > >
> > > >
> > > >
> > > > ----- Original Message -----
> > > > From: Graham Parkhouse <[email protected]>
> > > > Date: Tuesday, February 23, 2010 10:04
> > > > Subject: [Jprogramming] Agreement
> > > > To: [email protected]
> > > >
> > > > > I have a problem which is basically
> > > > >
> > > > >????? x v y
> > > > >
> > > > > x contains one sort of data and y a completely different
> > > sort. For
> > > > > simplicity let's say v has rank 0 and returns a rank 0
> > > result. I
> > > > > would like
> > > > > the result to have shape ($x),$y irrespective of the 
> values of
> > > > > $x and $y,
> > > > > but I cannot (easily) unless one is a prefix of the other.
> > > > >
> > > > > Why was this rule introduced?
> > > > >
> > > > > In my particular problem part of $y is pervasive through the
> > > > > process and I'm
> > > > > introducing additional structure through x.
> > > > >
> > > > > Regards
> > > > >
> > > > > Graham
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to