Re: Personal reply to Edd Dumbill's XML Hack Article wrt W3C XML Schema

[Jonathan Borden <jborden@mediaone.net>]

Richard Tobin wrote:
>
> The PSV Infoset as it stands is probably rather too closely tied to
> XML Schemas for this to appeal to those defining other validators (or
> rather type-assigners).  In particular the complex types and their
> representations are (of course) those of XML Schemas.
>
> Are the simple types sufficiently uncontroversial to form a basis for
> a common "typed infoset"?

Assuming:

1) an agreed upon XML serialization for the Infoset exists -- allowing
'infoset processing' to occur between perhaps disparate systems -- a
corollary is of course DOM representation of the PSVI information items.

2) Generally a standard mechanism or agreed upon convention for converting
between the type "xsd:foo" and a URI.

-Jonathan