• To: XML DEV <xml-dev@l...>
• Subject: Re: Article: "The horror of XML"
• From: "W. E. Perry" <wperry@f...>
• Date: Thu, 31 Oct 2002 10:50:11 -0500
• Organization: Fiduciary Automation
• References: <200210311523.PAA00170@m...>

Richard Tobin wrote:

> But if spec writers want it, they can always require an infoset
> corresponding to a well-formed XML document.
>
> If there was demand for it, the Infoset could be amended to include a
> definition of a "well-formed infoset", or something like that.

Well-formedness is a condition defined by lexical constraints, as in the
XML 1.0 Recommendation. The Infoset is distinguished from XML 1.0
precisely in that it takes an information-first view of XML entities, as
specifically contrasted with the lexical view of the Rec. As the tests
for well-formedness would be lexical, and applicable specifically to a
document rather than to a set of information items, definition of a
"well-formed infoset" would seem to fall between the stools of
'well-formed' and 'infoset'.

Respectfully,

Walter Perry

• References:
  • Re: Article: "The horror of XML"
    • From: Richard Tobin <richard@c...>

• Prev by Date: RE: Infosets a Horror Story? ( was Re: Article: "The horror of XML")
• Next by Date: unable to set external-schemaLocation property in XercesJ 2.2.0
• Previous by thread: Re: Article: "The horror of XML"
• Next by thread: Re: Article: "The horror of XML"
• Index(es):
  • Date
  • Thread