> > Tardis works like this.) For example: imagine a floor plan, at the middle
> > of which is a space the size of a closet. As you walk through the space in
> > VR, and enter the closet, you discover a Basketball court inside.
>
> We touched on this in chat last night. I think the metaphor will be a room,
> or for the formal, an office, or for those aging hippies, a "space." Each
> space will considerate itself the center of the universe.
That ought to satisfy a lot of folks. ;)
> One will hyperlink to another, but that could mean being handed off to a
> different coordination agent. Coordination agents will decide who is in
> the space and what sorts of "data stream URLs" need to hook up to.
>
> The quesiton you are asking (interpretted by me) is, How do the
> meta-coordinators work? If I say I want to be at X, Y, Z and there are
> difference spaces competing for that location, what happens? That probably
> depends on what space server you ask, no? I wish Mark would rewrite the
> Cyberspace protocol paper. (& sample code :-)
I understand. I also have another question. If you are standing in one
space, and look through a doorway into another space with different
coordinates, do you see anything? Does the other space only get displayed
when you actually go through the doorway? (I know this is implementation
dependent, but I am getting feedback that says you can't resolve the two
spaces at once in one view.
If we assume that links are only "buttons" that teleport you into the
starting point of the next space, then this is solved. There's another
question. You have to alter one space to make it point to another. If we
go back to the doorway paradigm, (even allowing that you can't see the
other space until you step through), how do you "park" your space next to
someone else's? They have to change the source code for their space, no?
So if someone creates Central Park, I can't park my Tardis in it. In other
words, I can't add space to another space's coordinate system. I can't
even park a normal helicopter on the roof of someone else's hotel
building. (Side-stepping the physical space coordinate-matching problems
of the Tardis, here.)
Yet this is what we will have to do when VR goes interactive. If physical
contraints are in effect, then more than ten users should not be able to
pile into an elevator at once, unless the server is altering the space on
the fly. If it is doing this, then it could also be translating the
coordinate systems. This would allow both the helicopter, and the Tardis
examples to work.
OK, so it's something which has to wait for the interactive stuff to get
done, so I'll drop the subject while we finish with the static spaces.
I think I'll go start building my Tardis.
---
Andrew C. Esh mailto:[email protected]
Computer Network Technology [email protected] (finger for PGP key)
6500 Wedgwood Road 612.550.8000 (main)
Maple Grove MN 55311 612.550.8229 (direct)
<A HREF="http://www.mtn.org/~andrewes">ACE Home Page</A>