Message text written by INTERNET:[email protected]
>
I think it important to distinguish between the complexity of the DTD and
the complexity of the streams that conform to the DTD.
<<<<<<<<<<<<<<<<<<
Scott,
Also notice that the DTD is generated from the XMI by slamming
everything into ATTLIST declarations.
This is NOT good for consistent information storage and retrieval
optimization in a database context. Sure the schemas will work,
but performance will go into the toilet as any information retrieval
will be forced to do a sequential search against the entire
information space....
So - it works to tranfer the model - but you cannot base the interchange
transactions themselves off the model instance.
Back to that glassblowing analogy - more research needed on optimizing
both the model and the interchange formats. An interesting rule-based
generation application - if you need someone to write some Prolog,
I volunteer!
DW.
