It runs "DxfToIv" and "IvToVRML" on the files - those are programs
supplied by SGI, and without source code I can't verify that they
generate nothing but valid VRML of course.
> I have noticed a
> LARGE number of so-called VRML files referenced from the VRML
> repository web pages are in fact Inventor files, and do NOT conform
> to the VRML spec.
Conformance is essential this early in the game. Looks like there is a
*strong* need for a "VRMLlint" - any takers?
> Is this why the InventorMentor is being summarised in HTML form - is
> the plan to quietly change the URL link from the original VRML spec to
> this summary document?
I'm as much a conspiracy theorist as the next guy (single gun theory my
ass! :) but I don't think we have to worry about SGI co-opting the
language development process, there's simply too much momentum from other
sides. Just wait until TCC's browser comes out :)
Obviously the way to ensure that one company doesn't railroad the process
will be to make the 1.0 specification an IETF RFC. The next IETF meeting is
in Stockholm, and not even I have an expense account that will let me travel
there, but I'll be at the Dallas one in December.
Before then, though, we can start whipping the spec into RFC language.
I'll see if I can get Roy Fielding to give me lessons in this :)
Brian
--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
[email protected] [email protected] http://www.[hyperreal,organic].com/