> I think programability might be beyond the scope of a markup
> language. I was thinking more along the lines of adding scripting to
> objects in a field of the object, rather than make all of VRML a
> "programming" language. VRML, at its simplest definition, should be a
> "scene-drescription" language. Maybe there should be something else that
> could be a fuller "scripting-language" that was a superset of VRML.
We want to model VR scenes as a collection of objects. VRML is used to
specfiy the objects' data while associated scripting languages are used
to specify the methods for the object classes. This allows you to cache
object data and methods *separately*. The issue at stake here, is the
formal model behind VRML that will allow you to specify data for novel
object classes without forcing those classes to be built into the core
language spec for VRML. My proposal will also allow you to syntactically
verify an arbitrary VRML "document" by checking against formal data
models for such novel object classes via network declarations at the
start of each VRML document.
-- Best wishes,Dave Raggett
----------------------------------------------------------------------------- Hewlett Packard Laboratories email: dsr@hplb.hpl.hp.com Filton Road tel: +44 272 228046 Stoke Gifford fax: +44 272 228003 Bristol BS12 6QZ United Kingdom