Re: request for guidance: centralized extensibility for HTML5 using X3D graphics

On Feb 23, 2010, at 2:17 PM, John A. Stewart wrote:

> Chris;
> 
> I don't know if you were around the Web3D Consortium when the X3D profile concept was created, but it sounds like it is what you desire.
> 
>> ... I think WebGL is a great first pass at what is needed. Any further steps have to be done with a minimalist approach or they will never gain any traction. The main point of my statements was that you don't need to give up functionality because of that minimalist approach.
> 
> I think you will find that the X3D Profile scheme provides a means of specifying subsets - Dr Behr's group at Fraunhofer (writers of X3DOM) and the rest of the X3D crew are trying to do just that.
> 
> Their "html" profile is at http://www.x3dom.org/?page_id=158 - it is certainly not cast in stone, but is a good start, IMHO.


I don't want to belabor this point too much, but let me be clear and specific about the current node set. The profile you mentioned has 73 nodes, which I consider to be just about an order of magnitude too large. 

The problem I see, and this is based on the 16 years of VRML/X3D's existence, is that having such a large node set makes it extremely difficult to produce interoperable implementations. By using the existing HTML facilities of event handling, scripting, DOM, image loading, etc. you reduce interoperability issues. By reducing the node set to the bare minimum (7 or 8 nodes) you reduce them further. The result is not X3D, but it is declarative 3D for the web that be used in the same applications for which X3D is being used today.

-----
~Chris
cmarrin@apple.com

Received on Tuesday, 23 February 2010 23:01:37 UTC