> Can you elaborate on "the DOM comments"?  What is the Expat parser?
>

I cribbed the DOM specification into the class and method comments.
The W3C says that you can crib it if you like, you just have to put
proper attribution.  If you notice, in the W3DomLevel1 class (I think)
there is an attribution, and all the comments have page references.
It seemed to me to be the easiest way to comment the DOM as it stands.
Perhaps I will want to go further later with comments and description,
especially in the case of items that are not currently standardized by
the DOM.

Expat is a parser written by the big SGML mogul James Clark, who is
the technical lead on XML.  Check out http://www.jclark.com/xml/.

The implementation is by no means complete, so don't think it is.  It
is a rough start, and I ran into unrelated trouble a little while ago.
I think it should be understandable where I am going, especially if
you have a look at Expat.  The goal is to process everything without
affecting the existing literal structure.  Of course, additions and
removals will affect documents, but I would like those to be
localized.  There are several reasons for this.  First, it makes sure
that documents can be handled verbatim, straight out of the parser.
Second, when working with extremely large XML documents, say, >1M, it
could make more sense to record incremental changes in a patch log
until a finished product is made.  This is facilitated by tokenizing
according to location in the document rather than by copying.

There are a number of things to discuss, and I'm going to pay
attention to the Swiki.

-John