I was influenced by the discussion in thread Peculiarity of Link, and opted for 
the uniformity of having every set satisfy 0 < L. .  I decided to see where 
that 
decision would take me.

In private correspondence it has been suggested that elements be required to 
satisfy [: -. 0 e. $ -- that is, no empty array may be an element.  What do you 
think?


Raul Miller wrote:
> On Fri, Jul 31, 2009 at 8:13 AM, Kip Murray<[email protected]> wrote:
>> As it happens, set box is the empty set.  It is a good thing that _nothing_ 
>> is
>> an element of set box .  Not allowing i.0 0 to be an element is the price I 
>> pay
>> for having an empty set in this model of finite set theory.
> 
> I do not understand why you do not use empty sequence
> to represent empty set.
> 
>    isset=: (-: /:~) *. (-: ~.@,) *. -:&'' +. 0 < L.
> 
> seems simple enough.
> 
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to