[sldev] Call for requirements: ISO MPEG-V (mpeg for virtual worlds) Deadline: July 16, 2008

[Tateru Nino]

Argent Stonecutter wrote:
> On 2008-05-26, at 00:06, Lawson English wrote:
>> OK, my dense brain finally gets it. I personally think that for 
>> objects, the primitives method is a better fit for a generic MPEG-5 
>> system than a generic mesh method, at least until bandwidth goes way up.
>
> That could be, but I don't think that anyone but Linden Labs will 
> support that. MPEG-5 would have to be something that everyone would 
> use, and everyone else is using a combination of meshes. The other 
> option (and one I hadn't previously though of) would be to just 
> encapsulate the game engine and then you lose the ability to 
> interoperate and might as well not bother calling it "MPEG-5".
The way I see it, it doesn't matter actually much *what* it is carrying. 
Ultimately we're talking about a transport protocol. Now, so long as the 
transport protocol is (a) bi-directional, and (b) supports connections 
to arbitrary network endpoints, and (c) is relatively legally 
unencumbered (requires no particular licensing to use or implement), and 
(d) relatively technically unencumbered (doesn't hamper or impede things 
to the point where there is no point in actually using it) -- well, then 
we'd be gold.