> whether you think of zero and one as numbers 
> or as boolean values.

0 and 1 are numbers; boolean values are numbers.
This is one of the tremendous innovations in APL/J
so taken for granted that it's usually not thought of 
as an innovation, but some people noticed.  See:

http://arxiv.org/PS_cache/math/pdf/9205/9205211v1.pdf
http://keiapl.org/anec/#implementers2
http://keiapl.org/anec/#Maple



----- Original Message -----
From: Devon McCormick <[email protected]>
Date: Wednesday, March 4, 2009 8:30
Subject: Re: [Jprogramming] LCM and GCD
To: Programming forum <[email protected]>

> A simple way to think about it is that "or" is based on "plus" 
> which gives
> you smaller results than "times" just as "GCD" will tend to be 
> smaller than
> "LCM".
> 
> More importantly, what is the GCD versus the LCM for boolean 
> arguments?  The
> result tables are the same as "or" and "and", 
> respectively.  This gives you
> consistent results whether you think of zero and one as numbers 
> or as
> boolean values.
> 
> On Wed, Mar 4, 2009 at 10:55 AM, Simon Jackson 
> <[email protected]> wrote:
> 
> > Hi just been reading the num denom posts, and just wondered 
> why the
> > LCM and GCD had the associations of and and or.
> >
> > I would have thought that if a factor is considered as an item 
> bit, then
> >
> > OR = LCM
> > AND = GCD
> > XOR = LCM/GCD
> >
> > The way that and = LCM and or = GCD is slightly counter 
> intuitive for
> > me, what is the reason for this choice?
> >
> > cheers confused jacko
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to