RE: [xml-dev] XML Binary Characterization WG public list available

["Dare Obasanjo" <dareo@microsoft.com>]

I'm both agreeing and disagreeing with Tim. Most applications deal with an abstraction layer that doesn't need some of the information items in the XML Infoset. However I'd translate this to most applications deal with an abstraction of XML that is similar or equivalent to some subset of the XML infoset. 
 
-- 
PITHY WORDS OF WISDOM
Blessed are the meek for they shall inherit the Earth, minus 40% inheritance tax. 

________________________________

From: Bob Wyman [mailto:bob@wyman.us]
Sent: Mon 4/12/2004 12:37 PM
To: Dare Obasanjo; 'Michael Champion'; ''xml-dev' DEV'
Subject: RE: [xml-dev] XML Binary Characterization WG public list available



Dare Obasanjo wrote:
> the XML infoset is too granular to represent the
> level of abstraction most real world applications
> deal with XML.
        I may be reading this wrong. Please help me here. It appears
to me that this could be re-worded as saying that "Infoset is too
rich..." i.e. you seem to be saying that Infoset is overkill? I could
understand if you objected to accomplishing the "smaller/faster" goals
by using a method that was "underkill" (i.e. important information was
lost) but I can't see why using an "overkill" method would be bad.

                bob wyman

-----Original Message-----
From: Dare Obasanjo [mailto:dareo@microsoft.com]
Sent: Monday, April 12, 2004 2:18 PM
To: Michael Champion; 'xml-dev' DEV
Subject: RE: [xml-dev] XML Binary Characterization WG public list
available


I tend to agree with Tim, the XML infoset is too granular to represent
the level of abstraction most real world applications deal with XML.
Most real world applications use an abstraction of XML that is more
akin to a subset of the XPath data model (elements, attributes, and
text nodes).

--
PITHY WORDS OF WISDOM
Blessed are the meek for they shall inherit the Earth, minus 40%
inheritance tax.