Re: ARCH: Virtual spaces and nets

Steve Hedges (cs0she@cis_server_1.sunderland.ac.uk)
Tue, 14 Jun 94 10:58:02 BST


Craig Presson writes:

> One project that some people will undoubtedly work on is building
> virtual copies of real places (cf. Gelerntner, _Mirror Worlds_). For
> this purpose, and for the advantage of familiarity, I recommend we begin
> with Earth as a shared space and allow multiple coordinate systems,
> such as lat/long/elev, geocentric Cartesian/polar/spherical, according
> to usefulness to the given problem.

I agree that we should try to maintain some link to the real-world
co-ordinate system, but we should recognise that people usually use
_several_ different systems, depending upon the size, distance and
type of the object being referred to. A discussion of this can be
found in "The naive physics manifesto" by Hayes (I think).

One possible approach is to have two co-ordinate systems - global
(which could be euclidean, spherical or even a ring topology) and
local (which would be specified independently by each developer). We
could see the global system as specifying the positions of a number of
rooms. Entering such a room would entert that room's co-ordinate
system.

Steve Hedges

Department of Computing and Information Systems
Sunderland University Phone: +44 91 515 2778
Priestman Building Email: cs0she@sunderland.ac.uk
Green Terrace
Sunderland UK