Re: Wording of ASOM

Quoth Rick, re AS section 18 description of WF:
>This is not worded well, because WF documents still MUST include
>internal entities and MAY include external entities (see XML 4.4,3).

Valid criticism, both because "DTD" can refer to the internal subset and
because XML did not officially define a processor type/mode in which no
attempt is made to use the external subset. Many implementations do offer
that as a nonstandard feature (or by deliberately creating a situation
where external subset retrieval will fail, eg by plugging in a SAX Entity
Resolver which no-ops that request), and perhaps the XML REC _should_ have
defined and named that behavior... but it didn't.

We can, if we want to. Maybe we should; there's clearly a demand for it.
But we'd have to come up with a name for it and spell out exactly what it
does; WF ain't it.

And if you really want to process without _any_ reference to the Document
Type, including its internal subset...  we can do that too but that's yet
another mode, and again we would have to assign it a name and precisely
define its behavior. (I'm not convinced this one is really desirable, since
it's hard to argue that the internal subset is not intended to be taken as
part of that specific instance document.)

______________________________________
Joe Kesselman  / IBM Research

Received on Wednesday, 16 January 2002 08:56:25 UTC