APPS & VRML target audience (rodney@odum.clr.toronto.edu)

Brian Behlendorf (brian@wired.com)
Wed, 13 Jul 1994 16:50:04 -0700 (PDT)


Rodney - you were listed as rodney@clr.toronto.edu - let me know if
you want me to change it so your posts don't bounce.

---------- Forwarded message ----------
Date: Mon, 11 Jul 94 09:04:11 -0400
From: rodney@odum.clr.toronto.edu (Rodney Hoinkes)
To: www-vrml@wired.com
Subject: APPS & VRML target audience

Hello VRML-people,

As a message was posted by Jason Cunliffe <jasonic@panix.com>
about some of the work my lab is engaged in (CLRMosaic in this case),
I thought this an opportune time to introduce myself and some of the
issues I would like to see the VRML definition approach:

I am a Landscape Architect by training and computer programmer
by interest for many years. I am the Head of Design Applications for
the Centre for Landscape Research, a research group focusing largely on
electronic media in urban and environmental planning and design. Our
primary exploration has been in the areas of interactive real-time
inquiry (for us real-time is a cognitive buffer swapping, falling in the
range of 0.25-1s). To date we have pursued this on Silicon Graphics
workstations but the advent of OpenGL will allow this to expand.

VRML is interesting as an attempt to bring 'VR' interaction
and spatial data into common practice in the Web. I have been pursuing
similar things in various forms for years but just last week I completed
an integration of some of this technology with NCSA's Mosaic. In doing this
I used YET ANOTHER format specification, one that has been tried and true
in our environment for many years. I mention this to bring some issues into
the VRML discussion that may not have been allowed for since there is not
a huge base of input from Architecture and Landscape Architecture (and
Planning) in these techo-discussions. These are:

* Support for arbitrary attributes on data objects for 'encoding' environments
ie. Landuse, Soil Pollutants, Material, etc.
a URL can be just another of these
- all viewers may not support querying these but the spec should
include this notion to be useful for REAL WORK DATA, rather than
several competing 'standards'
* Support for history of data - meta-data
ie Where did the data come from, Who created it?
At what accuracy??

The first of these is the MOST FUNDAMENTAL ISSUE in the ability of spatial
design and planning professions to use such a format and one that has kept
us away from Inventor (in at least perceived ways if not truly a limitation,
can someone confirm/deny this??).

CLRMosaic (see previous posting or check http://www.clr.toronto.edu:
1080/clrnew.html) is an attempt to integrate model data with HTML data as
BOTH ARE NEEDED - there will be little use if the two are not integrated.
I currently use an added HTML code of <MODEL SRC="file"> to include 3d
models. These are in our Spatial Data format but could as easily be in
other formats with the same method of inclusion. Our modelling software
pumps out the URL attachments from arbitrary attributes with it.

I hope the VRML spec will acknowledge these issues and support them.
If it does, I know our software will advance to test and support them but
otherwise a new 'standard' - a useful on to the professions I work with will
require development.

If you have an SGI machine - try CLRMosaic - it is in alpha right now
but useable - let me (and the group) know what you think.

Ciao,
-----------------------------------------------------------------------------
Rodney Hoinkes EMAIL: rodney@clr.toronto.edu
Head of Design Applications WWW-URL: http://www.clr.toronto.edu:1080/
Centre for Landscape Research AnonFTP: ftp.clr.toronto.edu
University of Toronto
-----------------------------------------------------------------------------