| Which brings up the debate: do we use a universally accepted coordinate
| system, or does each server maintain its own coordinate system?
[Todd Haedrich]
| The ability to make these links any type of object, allows us to break away
| from the concept of a room
Go to ISO 10744 - HyTime. Look at the concept of the FCS and FCSLOC.
(Finite Coordinate System and Finite Coordinate System Location... roughly).
The object in the FCS is independent of the FCS. FCSs are *not*
Cartesian. The quantum can be defined by the application. FCSs can be nested
while still maintaining a dominant frame of reference.
I was planning a post to this group on HyTime basics. I hesitate because:
1. Without a background in SGML, it is difficult to understand.
2. With a background in SGML, it is tedious to understand.
3. I cannot create an explanation of HyTime in VRML terms. It is designed
to be a general standard for integrated open hypermedia and must
be applied by the domain/subject matter expert. The VRML
group must decide if it is reasonable to apply and will be some effort.
The points: any technical presentation of it is long by the standards of this
group. Flames to that effect will be ignored. Understanding it is hard work.
The upsides: HyTime provides a grounding for VRML in an international standard.
The HyTime models are more powerful than the ones you are using, but you use
only the ones you actually care to implement. It provides declarations that enable a
browser to determine which features have been used prior to trying to
use the file. An upcoming corrigendum will provide a standard syntax for
defining node type (element type in SGML) class hierarchies.
Len Bullard