[
Lists Home |
Date Index |
Thread Index
]
From: "Mike Champion" <mc@xegesis.org>
> So what if XML were "refactored" so that the bare-bones well-formed syntax
> (and/or data model, that's another issue!) were the common core, and DTD
> processing were at the next layer up
Because it would not solve a basic problem (the same one that is endemic to any
post-processing model where the post-processing is not strict) that the user may end up
with a different information soup (i.e. the things available for Bill Perry to select his
information set out of) than the one the sender wants them to receive.
The other examples are XML Schemas, XInclude, and (because SAX only tells you
whether a document is standalone after you have already selected whether to use
a validating parser or not) DTDs.
At WWW7 I gave a little talk that we need to have a general mechanism for linking/binding
resources to elements/types/names ahead of getting a specific schema language; the
opposing Microsoft guy took the opposite side that because one schema language was
all that anyone would ever need, a more general mechanism was a waste of time.
Merely splitting up a technology isnt layering: for layering to work every layer must have a
mechanism to allow the next layer to be invoked or to fail if it is unavailable (as the default
policy.)
Cheers
Rick Jelliffe
|
