Re: ? universal name ambiguity

the descriptions of various mechanisms for typing, validation,
projection, are expressed in terms of universal names. while they are
expressed in terms of separate comparisons of the namespace name and the
local part of the name, that does not matter as, so far as i can
discern, the no semantics would change.

where, as the qualification "universal" would imply, a processing
mechanism - eg a query function, might be defined which is intended to
apply to more that one document.

the question is whether the xml community expects that these mechanisms
will always need to be contingent on the document descriptions which
were provided at the point where the query was defined, or whether the
community expects them to be general, and if so how that is to be declared.

as i recall, and from the present rec on XHTML read, the conclusion of
the three-namespace controversy, there are three document descriptions
for three variations of XHTML, all of which assert the presence of their
names in the same namespace, this although, for example, the type which
would be associated with the element 'p' is common among two document
definitions only, while the third one asserts its own type.

this would tend to make it difficult to define and locate elements for a
query library, since, in this case of xhtml, the namespace name does not
suffice to effect universal identification: the public identifier is
needed as well.

...

Received on Tuesday, 10 July 2001 08:31:26 UTC