Re: [Fwd: FW: "Roots" of confusion introduced at W3C]

["Joseph Kesselman/Watson/IBM" <keshlam@xxxxxxxxxx>]

Andrew Watt's point about terminology mismatches is probably valid.

Part of the problem here is that there are multiple legitimate definitions
of "root". The question is, root of _what_?

It may mean the root element of a document, which the DOM calls  the
Document Element.

Or it may mean the root of the tree -- in which case you have to specify
what kind of tree you're talking about. I believe XPath's definition of
root node corresponds roughly to the DOM's Document node.  But the DOM may
have detached trees whose (tree) root may be a different kind of node. And
some folks  have argued that XPath should be usable on subtrees/nodesets
other than full documents, which would require a similar concept of a
local/contextual root.

There may be other variations.

It seems to me that if we're going to reconcile these, we have to first
enumerate and describe what the real concepts are. Then we can figure out
which are and aren't similar enough to use the same name... and what names
to use. We may in fact discover that the term "root" really should apply to
more than one but has to be qualfied into "document root" versus "subtree
root" versus some other flavor(s).

______________________________________
Joe Kesselman  / IBM Research


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list