RE: [xml-dev] [Updated] A Categorization of Data Interoperability Problems

To start the argument:

It is a lot harder and more expensive to build ontological 
relationship systems than to require conformance to a schema 
based vocabulary.  As much as I am supporter of diversity, 
as anyone who has ever studied middle management knows, it 
is much more expensive in the middle to control both ends.

Open vs closed are extremes.  The common state is probably 
in the middle.   I would really hate to build a real time 
simulation client without a schema and an abstract object 
model that enables extensibility by definitions in the AOM 
rather than worrying about the encoding formats.

Most forms of interaction should be labeled and announced. 
Surprises are bad.

len


From: Roger L. Costello [mailto:]

A key point:
1. In an open system where there are interactions that cannot be predicted
apriori, it is unreasonable to expect a "standard format".  This is a key
point that I argue in "Living in a Schemaless Web":

    http://www.xfront.com/LivingInASchemalessWeb.html