On Thu, 3 Aug 2000, Simon St.Laurent wrote:
>  At 04:58 PM 8/3/00 -0400, John Cowan wrote:

>  >The Infoset *does* define abstractions for XML.  It doesn't define
>  >every possible abstraction for XML.

>  More to the point, it doesn't define abstractions for all of XML, leaving
>  them as 'fix in documentation' for those who need more than the Infoset
>  currently provides. As the Infoset itself is basically documentation, this
>  doesn't seem reasonable.

Why not? The infoset seems mainly to be a guide for (W3C)
spec-writers, saying "unless you can think of a good reason otherwise,
your spec should deal mainly with /this/ level of abstraction of XML
documents". Most W3C specs (DOM, XPath) do operate at more-or-less the
same level of abstraction, so using the infoset just takes the
"more-or-less" bit out and ensures that most future specs operate at
/exactly/ the same level of abstraction. This seems to me to be rather
useful.

-- 
Richard Lanyon (Software Engineer) |     "The medium is the message"
XML Script development,            |             - Marshall McLuhan
DecisionSoft Ltd.                  |