Just to help me, as part of the VAG, get a feeling for what people on
the list want (which I assume will be very contradictory), could
people please send me mail about where they would like the line drawn
between VRML 1.0 and the next release? Bear in mind that the more
goes into the next release, the more time it will take to produce it.
I.e., a very simple version of 1.1 could probably be released in the
next day or two, but it might not have very much extra stuff, and full
behaviors in VRML might take a month or two more of debate to esach
closure (just in the VAG).
Here is a list of features, ranged in rough order of difficulty. Let
me know where you draw the line, and where is the minimal useful next
esvision of VRML:
1. minor clarifications/simplifications to the VRMl 1.0 spec
2. simple new nodes: i.e., terrain, environment, generalized
cylinders, etc
3. user/geometry collision detection
4. multimedia types: audio, video, animations
5. triggers and sensors, with simple wiring (walk into the room, the
light turns on)
6. External API to allow for extensions (Java, Telescript, Perl,
whatever)
7. Fully programmable VRML language (inheritance, encapsulation,
compositing relationships, etc)
8 VRML-based networking protocol, allowing for multiple
participants.
As for me, I think that VRML 1.1 falls somewhere in the vicinity of
4-5. And VRML 2.0 falls somewhere around 6-7. But other people have
different plans for VRML, and I'd like to hear their own priorities,
and triage list for what should be part of those two standards (or
even if there should be both a 1.1 and 2.0). If we're all willing to
wait another month or two, I think 2.0 is becoming very close to
resolution, and we may be able to jump straight to it.
Comments/questions/concerns?
Tom Meyer