Re: 3D -- WebGL and SVG?

On 03/08/2011, at 9:24 AM, David Dailey wrote:

> I had a brief interchange with Neil Trevett [1] of the OpenGL (Khronos) group recently in New York. As you know WebGL is pretty mature as a spec now and has good traction in a number of browsers. One writes very low level code dealing with 3D shapes, textures, shaders and the like. It is pretty cool stuff, as I’m sure most of you are aware. It is not exactly, though, what we might call “tractable” to the masses. Neil said something about their being tens of thousands of 3D graphics programmers and millions of web developers. I asked about possible work with the W3C and he said that there is an incubator group considering such things. One of the obvious tricks is to design a markup language sophisiticated enough to handle all the wonderful stuff that WebGL would do under the hood. I suggested that the SVG WG might be interested in such collaboration. No doubt some of you know about this already – perhaps all of you do – but the topic is certainly of interest to a broad audience. Incidentally, <replicate>[2] gives one of the most concise grammars I’m aware of for describing, in human-friendly terms, a very rich class of 3D-structures.
>  
>  How else might we conceptualize markup for 3D? Is this something worth thinking about here?

It depends what you mean by "thinking about". I personally don't feel that the SVG WG is the right place for discussions on 3d. I also think that markup languages are extremely difficult to get right (considering there is only one truly successful markup language at the moment), and that 3d has such broad use cases that at best you'd be targeting a small subset of the community. Speaking of community, the W3C has very little involvement from there - I can't see how it would manage anything better than X3D or Collada, both of which were designed by domain experts over a long period.

The only thing that's in the domain of SVG is making sure that it can be used with 3d technology. I don't think there is much to do there, but I might be wrong.

Dean


> [1] http://en.wikipedia.org/wiki/Neil_Trevett
> [2] http://svgopen.org/2010/papers/46-A_proposal_for_adding_declarative_drawing_to_SVG/index.html