Peter Amstutz wrote: > Since most of the work has gone into the network layer, a lot of stuff > in the 3D layer was a quick proof of concept rather than being really > well thought out. Even if s4 -> s5 transition wasn't necessary, an A3DL > redesign was already on the plan.
There's a lot of unwritten stuff like this for a newbie to the VOS project like me :) But it's good to know. I didn't really know how set you guys were on the current state of a3dl, but how much it seemed like "make it work easily in CS" was the main design objective made me a bit uncomfortable with it. But knowing that it was mostly a proof of concept and that a redesign is on the plan makes me feel a bit better. So, I don't know if it's too early to start talking about redesigning a3dl, but thinking about it today I had a thought. You guys already want to have VRML (and eventually X3D) translation to and from VOS/Interreality 3d data. Well, why not make it easier by desigining a3dl to be as close to 1:1 with X3D as possible? The X3D guys are putting a lot of work into hashing out the things that an interoperable 3d standard needs, so why re-invent the various wheels? I'm not talking about adopting X3D file formats as the interchange format of course, but rather having the object types and properties have a 1:1 correspondance with those in an X3D scenegraph, and have the formats for textures and meshes and shaders and such be X3D compatible. Of course, it doesn't have to implement *all* of X3D, and will have its own extensions on top of what X3D specifies (I'm thinking something like inter-server portals would probably be very vos-specific). But it might be a good idea to ride on top of their basic design. Also, doesn't X3D specify different optional modules and levels of compliance? You could have that kind of information published by servers and clients, even, to help with service discovery. But I actually don't know too much about X3D at the moment -- learning it is on my ever-expanding todo list :) There will probably be some extra work in making crystal space loaders for a3dl if it's done this way, but it's probably better for the VOS standard in the long run. Of course, if there's some reason you guys don't want to do an X3D-inspired design and think it'd be better to start from scratch with the 3d scene data I would be really interested in your opinions. Does X3D have any serious flaws that would hold it back from being the "right" data model for Interreality 3D? -Ken _______________________________________________ vos-d mailing list vos-d@interreality.org http://www.interreality.org/cgi-bin/mailman/listinfo/vos-d
