Physical Space Representation

[Mathue Moyer]

This is the first time I've posted to this list since being invited, so
recent trends indicate that I ought to introduce myself.  Unfortunately,
in the context of internet game development I'm pretty much nobody when
compared to others on the list, so I won't bother you with the details.  :)

I've been giving a lot of very undirected thought to the notion of Physical
Space Representation over the last few years, and have dreamt up and discarded
a fair number of schemes.  In reading the recent messages on this subject,
a cloudy memory from one of my very first programming classes reared its ugly
head.  I seem to recall some pretty slick algorithms for handling sparse
arrays, and I wonder if people have considered using them in a PSR strategy.
I really wish I could find my notes on sparse arrays, but the gist of the
idea is that you can provide efficient (in time and space) mechanisms for
representing and manipulating very large arrays (don't recall if it handled
 3 or more dimensions) in which most entries contain no information.  This
seems applicable to many traditional PSRs in which you have isolated clumps
of interesting data and lots of irrelevant stuff flushing out the rest of
the world.  If this sounds vaguely interesting to anybody, or if anyone
happens to know off hand of a source for sparse array algorithms and data
structures, do chime in.

On a slightly different note, I think it would be somewhat interesting to
catalog (in a fairly cursory fashion) all of the PSR schemes people have
come up with, heard of, etc...  Seeing one list of the whole lot (or at
least a decently-sized subset) could help generate new ideas.  Here's my
first stab at categorizing PSR attributes (off the top of my head, I confess)...
I hope the whimsical acronyms aren't too hard to follow.  ;)

Object Interaction Granularity:
   Within a single Frame Of Reference, how do you determine whether or not
   objects can interact in various ways?  Traditional room representation has
   very low granularity.  Objects are contained in cells, all cells are
   of equal effective size, objects can only interact (forget about "tell")
   with each other if they are in the same cell, and all methods of interaction
   are available as long as the participants are in the same cell.  Absolute
   coordinate systems, on the other hand, have very high granularity.  Objects
   are tracked by distance in one or more dimensions from a reference point.
   Interaction is based on distance between objects, and different methods of
   interaction can easily require different ranges.

Frame Of Reference Structure:
   How many FORs exist in the PSR and, if there are more than one, how are
   they related?  Traditional room representation and absolute coordinate
   systems have a very simple FOR structure: there's just a single FOR.
   Other PSRs allow adjacent FORs (e.g. the ocean FOR borders on the forest
   FOR borders on the town FOR etc) and/or nested FORs (multiple adjacent
   dimensions which contain universes containing adjacent solar systems
   containing planets containing adjacent land masses and oceans...)

Hmm... apparently I'm finally awake, because I've run out of philosophical
hogwash for the moment.  I imagine motion of objects might be a qualifiable
attribute, but I can't figure out how to divorce it from OIG and FOR
structure yet.  Feel free to extend (or ignore) the madness.  :)

mathue

--
       Mathue Moyer                           | Anyone who says they have only
Email: mathue at cts.com                         | one life to live must not know
URL:   http://www.users.cts.com/king/m/mathue | how to read a book.