>The problem is that if the parser ignores the DTD, how can it
>detect #1 above ?
Obviously it can't. If a parser wants to fully validate an XML document it
has to read the entire DTD. One of the things it must validate, if the
document has RMD=IGNORE, is that the DTD could be ignored without changing
the data the application received. A parser that is not validating, on the
other hand, can choose to take advantage of the RMD decl and not parse the
DTD. Provided that the document has been validated, the non-validating
parser will be guaranteed to get the same results as the validating parser.
James