[
Lists Home |
Date Index |
Thread Index
]
From: "Simon St.Laurent" <simonstl@simonstl.com>
> Least disruption seems like a reasonable approach to me.
I agree.
A further point is that the human use of an XML document as a marked-up
text is not always using an "XML processor" in the sense that
the spec uses. An XML processor parses the XML
and makes available information items (see Appendix B
http://www.w3.org/TR/xml-infoset/ ),
but an XML editor may not even parse the document, or may parse
it in some devious or incremental way: an editor or reader of
an XML document may work off the text of the entity not the
infoset.
Contrast applications which work off text with applications that
extend the infoset: such as XML in canonical form where the
attributes are given in alphabetical order: in that case, I suppose
the basic infoset is notionally augmented with positional indicators.
Cheers
Rick Jelliffe
Orlando, Florida
|
