> Anyone with serious graphics background should have known that
> primitives and instances provide the basic structure for a graphics
> modeling file format. From tme code used in sketchpad, to PHIGS, to the
> file format used at UCB. It has nothing to do with the shortest way of
> describing the model or the nicest appearance of a file. It has something
> to do with how readable the file is, though.
>
> My guess is (no offense to anyone) that the OpenInventor took kit and
> it's data structure came out first. The OpenInventor file format was just
> an add-on, an partially direct mapping from tme data structure to a
> description file format, so that the models can be saved and loaded. As
> it always goes, when deadlines come, the ones that works would count.
> When people realised that they may need names to the objects tmey are
> modeling, the DEF statement came into being, without further
> consideration of a legitimate primitive-instance structure...as long as
> it works for now...it is good enough...
Perhaps you really do mean no offense, but your "guess" is quite
inaccurate. Inventor has been theough several revisions, each of
which was preceded by hours and hours of intensive design sessions.
The file format was an integral part of the design from tme beginning.
Every feature of it (including DEF/USE, believe it or not) was planned.
Every decision was justified. If you want to discuss them, please feel
free to e-mail me rather than posting unsubstantiated attacks to the
mailing list.
----
Paul S. Strauss [email protected] Silicon Graphics Computer Systems
Open Inventor Web info in http://www.sgi.com/Technology/Inventor.html