| Boy did you say the wrong thing in the wrong place.... :-)
Perhaps, but he is expressing a real opinion founded on his
viewpoints, so it is worth a listen. I have a similar concern,
but not about the "doability". I've seen sims with behaviors
and it is quite doable once y'all agree on how. My concerns are more
of timing and precedence. Teying to get two VRML browsers to do
something similar with the same file is a very dicey proposition
with VRML 1.0. This is not good. To the lay user, this means
VRML fails to provide interoperable 3D scenes across WWW
systems. This, in effect, means it fails to meet its primary goal
at this time. This may be temporary, but it is the perception
and in a sensual world, perception is king of the opinion stack.
I realize the newness of the spec, the betaness of the freeware,
and all that. I am entirely grateful! I am also, like many others,
sitting on a lot of executables that don't yet deliver on the promise
of VRML 1.0. Yes, graphics is harder, larger, and targeted to more
sensations, but where the dominant sensation is sight and the
dominant use is navigation, the text systems still work better.
Flame if you must, and I also understand the difficulty of building
a House of the Future Together, but we must honestly stop
and ask ourselves where we are at the moment, and where we
can get in the next cycle.
Is it possible that the system vendors are engaged in a
ferocious if polite competition to get to the market first
with the most capability in order to make their own
extensions the *standard* ones? If so, the VAG has
a real bear to wrestle. Sell tickets and let us watch. ;-)
We need a baseline of stable systems. The HTML browsers
are degrading gracefully when some NuTEscape extension is
added. If you expect us to homestead this environment, at
least the browser can't just die, or say "ERROR!". Even if
as in WorldView, it simply says where to look to find the
problem, it must do something to move the user forward.
BTW, Tony, et al, is it possible for it to suggest a corrective
action like "Remove ??? node and replace with ???" It doesn't
matter if things slow down if the file has to be edited anyway.
Also, if it kicks off even an ASCII editor, that would be nice
until graph-driven editors are available
The browser programmers for VRML are probably worked to
death right now, and you all have my sympathy, respect, and
best regards. IMO, a season for focusing on VRML 1.0
is needed, not 2.0., at least until WebFX, WorldView, WebSpace,
Fountain, VRScout, et al can reliably exchange 1.0 files. (It
would be nice if an HSB export could play on anything!)
So on with the debate, but can we get our expectations
in line with our schedules? I'm just a neophyte user, yet I
would like to get on with the art of this and do my part to sell VRML to
customers. However, all I can tell them right now is,
"it's warming up in the bullpen. Expect a shutout." Some
customers will be very happy right now with a part catalog that will
rotate and zoom on parts, and exchange link information
with an SGML/HTML browser if the system is reasonably
costed, and impervious to uninvited buggery.
Perhaps a tmesad on exactly why such interoperability is
not happening would be good "leaven on the waters" right now.
Which features of VRML 1.0 are the most contentious
for the programmers?
Len Bullard