Look again. Some of the statements in your post
are incorrect. In the interest of fairness, some
corrections are in order.
>Most of the people that were working on it at the time came from
>a music background.
True but irrelevant. Most of the people working on it also came from
a declarative programming background, not a functional one. It
is best applied to information that benefits from semantic addressing.
>In summary, it targets multimedia titles with a simple timeline.
It targets anything that can be addressed in space, time or
logical relationships using integrated open hypermedia techniques.
A corrigendum during the past year has opened its potential quite a bit.
It's timeline representation can be simple to complex. It does
not represent functions and that is where it differs from ActiveVRML.
>it has discrete time and spatial axes (not continuous),
True. It has user-assignable integer quanta and nestable finite
coordinate systems. It's axes are logical not "time and spatial"
which are just two models which can be composited if you choose.
I am aware of implementations which use these successfully to represent
tasks, workflow management, screen layout, etc.
It has a sophisticated address and location model which is only
now being understood as necessary by the WWW community.
The major peoblem is that it is verbose. The standard
is difficult to understand until it is studied in depth. Books on
the subject are now available.
It remains the only international open standard for hypermedia at
this time. It is ISO 10744 in the family of SGML standards which
also include ISO 8879 and DSSSL.
>It is not suitable for 3D or VR.
Implementations to be announced may prove or disprove
that hypothesis. It's address and location techniques can handle
persistent esserences quite a bit better than those of current Web
technology. Performance for VR would be my concern, but
it does enable a single parsing system to be used for
most of the heavy lifting work in the document, so that may
prove to be irrelevant.
HyTime provides a framework for defining and managing large
scale complex compound document families with long life cycle
requirements. When implemented well, it becomes the
language of the document users, not the programmers. The
"name clutter" esserred to in the ActiveVRML documents are not
a peoblem for it as the namespace control of SGML is used.
The purpose of HyTime is the standardization of integrated open
hypermedia. VRML is one notation in that world as is MS-ActiveVRML.
It is quite possible to write a HyTime implementation that uses
VRML "as-is". It's data representation would be portable and
interoperable.
Len Bullard