Len writes:
> It is said, the VRML as it is is too hard. It is said
> that the Web content community has not embraced it.
> Ok. This list has most of the talent that has embraced it.
>
> o What can't it do that it should to make applications
> easier to create?
It should be much easier to implement. That inevitably means
simplifying the language itself, to remove those features
that create a needlessly high barrier to implementation.
If a company wants to implement a plugin that's maybe
200k in size that does streaming real-time 3D, they cannot
use VRML -- period. The smallest VRML browsers are way too
bulky.
They could try to leverage existing VRML browsers, but there's
no really efficient way to do streaming into them. I know that,
because I've done it (for VRML Dream). It's hard to get working,
it requires using *both* the EAI and JSAI, and even then it's slow
compared to a custom solution.
Another issue is that many companies need to protect their intellectual
property, since that's their bread-and-butter. VRML doesn't have any
way of doing this. Anyone could theoretically steal the content.
Yet another issue has nothing to do with technology, but with perception.
Companies who have been tracking VRML's progress have seen that there
was a big discontinuity in going from VRML 1.0 to VRML 2.0, and they're
about to see another big change in going to X3D. They've also seen
a lot of VRML-only companies fail. Technical issues aside, those facts
alone are enough to convince many companies that a proprietary solution
is going to be more stable. Fixing the technology isn't enough -- we
have to somehow build mindshare.
> o Given some range of applications (you choose), what
> components do you need to build them if the geometry
> component is common?
Streaming. Both streaming of animation data and streaming of audio,
which can be spatialized using the equivalent of Sound and AudioClip nodes.
--
Bernie Roehl
University of Waterloo Dept of Electrical and Computer Engineering
Mail: [EMAIL PROTECTED] Voice: (519) 888-4567 x 2607 [work]
URL: http://ece.uwaterloo.ca/~broehl
