> Mind, I'm none too happy with SGI/TGS at the moment, either --
> contrary to their denials, the reports seem to be indicating that
> they *are* pulling a Netscape, whether intentionally or no. That's
> an important issue, which we need to start hammering on now,
> because while I'm moderately comfortable with the notion of
> inheriting most or all of Inventor's *static* properties, I'm
> not *nearly* as sanguine about picking up its *dynamic* ones.
> 'Course, I'm not at all certain that the motion commands in
> Open Inventor really *are* dynamic in any meaningful way, but
> that's an argument for another list...
Personally I have a very different oppinion of the situation. I have
been in close contact with folks at both SGI and TGS, and I have been
very pleased with them. Both are very excited about VRML and have been
very good about trying to keep a clear separation between VRML and
OpenInventor. One thing is also very clear, they are not trying to pull
a Netscape - both are very converned about not letting the VRML
standard splinter off into lots of pet extensions. For instance,
Webspace is completely VRML spec 1.0 (as currently net-published)
compliant. All of the extensions they have added are completely
compatible with VRML. A good example is their "vantage point"
extension. It is a switch node called cameras that their browser can
recognize and allows you to switch the camera view from a menu in their
viewer. This allows them a nice function, and any browser that doesn't
recognize the extension will see the default witchChild of the node.
Len Wanger -- [email protected]
Interactive Simulations Inc.
http://www.intsim.com/~isigen