At 09:46 -0700 1999/07/20, Carl R. Witty wrote:
>For these new lists to be computationally useful, you need new
>operations (more than the "check for empty list", "take the head", and
>"take the tail" provided by Haskell). What new operations would you
>suggest providing? How would they be implemented? (Would they
>involve increasing the size of an evaluated list object, to add more
>pointers?)
These are good questions. Peter Hancock and I have in private
communications worked through some ordinal theory to check for the basis of
a mathematical theory of these "ordinal lists". Perhaps we can return on
the issue at some point.
Probably good examples will lead the way. One variation could be that one
knows how to compute the index value of a list, and that the list comes
with sufficient recursive information (including limit ordinals) enabling
such an evaluation.
If somebody has some good suggestions, let's hear about it.
Hans Aberg
* Email: Hans Aberg <mailto:[EMAIL PROTECTED]>
* Home Page: <http://www.matematik.su.se/~haberg/>
* AMS member listing: <http://www.ams.org/cml/>
