Arrgh! - FW: Call for unifying and clarifying XML 1.0, DOM, XPATH, and XML Infoset

[Nils Klarlund]

> >1. The Infoset is not the same as the DOM. The Infoset does not imply a
tree
> >structure. The Infoset should never imply a tree structure, as it's
intended
> >to describe the behavior of non-node-oriented processors such as SAX.

  "This specification presents the information set as a tree for the
   sake of clarity and simplicity.."  (verbatim, from Infoset, Last
   Call).

"Tree" is a widely used term among computer scientists, logicians,
mathematicians, and so forth.  The quintessence of XML is that it's a
linear syntax for labelled trees.  (Other characterizations are
possible, of course.) The tree view is the key to the superiority of
CSS and XSLT as an abstraction mechanism over text-macro based markup
languages (like TeX/Latex).

>> I believe the W3C made the right call by making attributes not
> >children of the elements with which they are associated.

Quite possibly, but other nodes deserve some nomenclature as well, not ad
hoc
treatments.

> >3. The term "information item" is a necessary one because of the overuse
of
> >other synonyms (such as element, attribute, and so on) in the DOM
> >specifications. Information items are none of these.

Would you call page numbers "page information items"?  The information items
are nodes in a tree (with ordereded and unordered children) and should be
called
nodes.
|
> >4. Nils appears to be obsessed with terminology in the XML 1.0
> >specification, which is significantly out of date at this point. The
newer
> >drafts should be taken as the "official" W3C position - there's just not
> >anyone providing errata to the original document to keep it in line. For
> >example, the current way of thinking says that there's a Document node,
> >which must have exactly one Element child node. XML 1.0 says something
> >slightly different, but that's not important.

No, I'm concerned that DOM, XPATH, and INFOSET do not turn out to be
essentially the same model.  That would be a great shame.  Better get it
right
than rush.

/Nils


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ or CD-ROM/ISBN 981-02-3594-1
Unsubscribe by posting to majordom at ic.ac.uk the message
unsubscribe xml-dev  (or)
unsubscribe xml-dev your-subscribed-email at your-subscribed-address

Please note: New list subscriptions now closed in preparation for transfer to OASIS.