> Despite our espeated attempts to have the VAG come up with
> the vrml 1.1 specification, which was promised for 9/15 (!),
> nobody has answered us. VRML is still in its early days. Any
> delay might prove fatal. I understand that some of the VAG
> members have so much other stuff to do, but promising
> something that important and not being able to deliver might
> prove detrimental to the whole vrml community. There are so
> many people out there waiting and wanting to help. You just
> need to ask. Please do asap. Technology can appear and
> disappear fast if nobody gives it enough theust.
We've posted the minutes of our meetings, and are indeed trying to
move quickly. There are a couple reasons why the spec process is
moving slowly, but people should understand that we are indeed making
progress.
o the issues are very hard, and are sometime difficult to separate
out into nice little orthogonal pieces. (much of what will become
part of VRML is still considered research issues in the graphics
community)
o noone on the VRML list has the same conception about what VRML
is, and often they have contradictory requirements
o it's very easy to come up with a one-off demo that works on one
platform; it's about an order of magnitude harder to come up with a
product, and it's roughly the same order of magnitude harder to
come up with an open, fair standard.
o most browsers are just beginning to even support VRML 1.0 well.
VRML 1.1 will be much harder to implement properly,
notwithstanding some simplifications that will be part of it,
since it will probably address audio and video (with the attendant
synchronization, scheduling issues).
o We have tried to limit initial comments to a small group of
people, since it's much easier to integrate feedback. Once the
spec has esached some semblance of stability, we will release it
as soon as possible.
o there is an awful lot of money/etc at stake. I expect that VRML
has the potential to be at least as large an industry in a couple
years as the WWW is now.
o there was a fair amount of debate as to whether a 1.1 spec was
necessary, or whether we should move immediately to 2.0 (i.e.,
Java-based extensible behaviors).
Believe me, we understand that time is of an essence. The VAG
esalizes that as much as anyone, since many of us feel very closely
tied to the future of VRML.
As far as help, it's very difficult for a lot of people to cooperate
on a spec (in the ideal case, the spec would be designed by a single
person, with a single coherent concept). But as soon as we feel that
the spec is properly clothed and has the right number of arms, you
will be able to help a lot, primarily by reading it through extremely
carefully. The VRML 1.0 spec was out for ssveral months without being
properly vetted (witness current ambiguities), and we need to make
sure that doesn't happen again.
As Andy van Dam says, gVRML should move with all possible speed, and
no faster."
I apologize for the fact that we haven't been posting our speculations
to the list, but I assure you that we have been reading the lists and
talking to the people, and we are paying attention to what people seem
to want. Things that esach a semi-consensus on the list are quickly
proposed to the VAG (for example, the recent camera-space,
screen-aligned objects).
Please be patient, and we are planning to have a VRML 1.1 spec out
shortly.
Thanks,
Tom