As a start on Raul's problem below, would you be willing to use the following 
for-this-problem definitions?

A "sequence" is a list of boxes.  A "term" of a sequence is the open of one of 
its boxes.

A "set" is a sequence whose last term is i.0 0 .  An "element" of a set is a 
term of its curtail.

The set-marker <i.0 0 tells you to treat sets differently from sequences. 
Sequences s and t which are not sets are treated as the same if (# s) = (# t) 
and term k of s is the same as term k of t for k e. i.~ s .  But sequences H 
and 
K which are sets are treated as the same if no element of H is not an element 
of 
K, and no element of K is not an element of H.  If one of sequences s and t is 
a 
set and the other is not a set, they are not the same.  If array a is not a 
sequence or array b is not a sequence, they are the same if a -: b .

I would be very interested in verbs which test whether two sequences are the 
same, using these definitions.  Raul please recast the problem if this 
formulation is too far from what you had in mind.

Kip


Raul Miller wrote:
> On Tue, Aug 4, 2009 at 2:37 AM, Kip Murray<[email protected]> wrote:
>> I shudder at the task of figuring out what the task is, and opt for 
>> appropriate
>> "sort nubs" (Fraser's verb Set) as the argument set is being created and as 
>> the
>> value set is being created.  Try to use operations which _preserve_ the 
>> property
>> of being recursively sorted and nubbed.  (_preserve_ as contrasted to 
>> _impose_)
> 
> Here's an example problem which might help.
> 
> You are working in an application domain which uses both sequences and sets.
> Sequences and sets may both contain other sequences and other sets.
> 
> How do you tell if two sequences are equal?  How do you tell if two
> sets are equal?
> 
> (There is more than one way to implement this system, but some
> approaches will be simpler than others.)
> 
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to