On Jun 4, 2009, at 11:23 AM, mark richardson wrote:
Hi Torsten,
Torsten Anders wrote:
If you know all the features of your data structure in advance, then
why not using plain (determined) records?
I am using objects in constraint problems all the time -- using them
in a script is strait forward. The advantage is that you can define
your data structure incrementally (inheritance). The disadvantage is
that it takes more memory if the objects are in the script and are
therefore copied all the time during search (when their spaces are
copied). So far I have not found a way to have the objects outside
the
script and only the variables encapsulated in the objects are inside
the script (or at least no way without much programming overhead
effort).
Thanks for your reply.
Yes, I can use standard records for the initial state of my virtual
actors, what I can't get my mind around is how to model the change of
state. I'm comfortable with the ideas behind FD and FS, including
where
part of the domain may be already determined. But I haven't taken
constraints any further until now.
What I'm not comfortable with is that my initial state for a virtual
actor needs to (possibly) be changed. In other words I'm using
constraints to alter information rather than search for the
'missing bits'.
I think the problem I'm having is that I'm trying to find a data
structure that will hold the initial state *and* allow changes via
constraint based search to achieve a resultant state. I'm now thinking
along the lines of a standard record for the initial state and a
separate 'RecordC' for the resultant state.
Does this sound more sensible?
Constraint problems are always stateless. You can add information in
the sense that some piece of information gets more specific (e.g.,
turn {1, .., 10} into {2, 3}). However, you cannot change state.
Best
Torsten
Regards
Mark R
--
Mark Richardson
Research Assistant
University of Teesside
[email protected]
[email protected]
______________________________________________________________________
___________
mozart-users mailing list mozart-
[email protected]
http://www.mozart-oz.org/mailman/listinfo/mozart-users
_________________________________________________________________________________
mozart-users mailing list
[email protected]
http://www.mozart-oz.org/mailman/listinfo/mozart-users
