As the Web3D Consortium Turns

Dateline: 3/1/00

At the Web3D/VRML 2000 conference the big news of the week was the (yet again) reorganization of the Web3D consortium. A week earlier a minor scuffle broke out on the "big list" (the mailing list) concerning the proposed limitations on participation in the Consortium Working Groups. The "community" got pissed off that the proposal was to limit participants to members of the Consortium. This was coupled with the new "feature" whereby the Consortium was offering individual professional memberships. It was a nice try on the part of the Consortium but it wasn't accepted. Fortunately, and to the Consortium's credit, the actual policy is to let anyone participate in the working groups continuing the current practice. Membership, however, allows one to participate in various types of voting. There are three classes of votes. Only the top level dealing with consortium policy, not the technical stuff is closed from individual members. Specifics should be up on the Consortium web site soon.

On the technical side there was also a major controversy that's been brewing for a while concerning the future direction of X3D. First a little background on X3D.

X3D, the name, was created primarily as a marketing mechanism. For a variety of reasons VRML, has become a tainted technology in the USA. Actually in Europe VRML is viewed as cool and the way to do 3D on the Web for a wide variety of applications. (personally I think the Europeans are a step ahead in this issue) VRML never lived up to the original hype, what technology could? However VRML also was problematic and implementations like CosmoPlayer, and WorldView were incompatible with each other. More importantly the VRML spec itself has significant problems. No browser could be "compliant" to VRML because such a thing is impossible. Content that plays in one browser might be "conformant" to the spec and behave and appear quite different in another browser all are "compliant". Ambiguities exist and undefined behaviors are widely acknowledged in the spec itself. This is not due to a lack of foresight on the part of the browser builders or the creators of the standard but is most likely due to the great rush to get VRML out the door, without enough implementation experience. But that is all old news.

X3D is widely perceive to be an attempt to encode VRML using XML syntax. While this is true it's also one of the least important aspects of X3D. In reality X3D is the next generation of VRML. It will (we hope) fix the ambiguities and undefined behaviors of VRML. It will allow much simpler forms of extensibility and will integrate well with future Web technologies (via the XML encoding). If all goes as planned it will be possible to load small compact X3D browsers that can incrementally load "profiles" which perform one or another functionality, such as compatibility with VRML 97. Over the past few weeks a technical issue concerning the X3D DTDs has arisen. (A DTD is a Document Type Definition which defines the structure of a document, in XMLese. You might also think of it as a data base schema ...sort of) A possible technical solution to the issues regarding X3D DTDs is explained in Composing Scene Graph Alternatives on the Consortium's web site.

Now that the technical issues are getting resolved (famous last words), what else happened at the big Web3D group grope?

Next page Other Web3D Happenings >Page 1, 2