On 1/30/08, Daniel Aleksandersen <[EMAIL PROTECTED]> wrote: > > But my question on why the atom:id element is being used instead of the > xml:id attribute still stands. >
It is desirable for the identifier to be an IRI. You can ensure global uniqueness this way and your identifiers can be structured, readable, and possibly participate in existing identifiers used on your web site (such as the entry's permanent URL). While you might be able to adopt additional constraints to apply to xml:id attributes to achieve the same goals, this would be awkward given xml:id's restrictions, potentially confusing since you're effectively overloading the semantics of xml:id, and potentially painful if someone attempts to create or embed an xml:id obeying XML's rules but violating Atom's rules. Not all XML within an Atom document will be Atom. David
