RE: XML observation

> -----Original Message-----
> From: ext Martin Duerst [mailto:duerst@w3.org]
> Sent: 08 July, 2003 16:50
> To: pat hayes; Stickler Patrick (NMP/Tampere)
> Cc: w3c-rdfcore-wg@w3.org; w3c-i18n-ig@w3.org
> Subject: RE: XML observation
> 
> 
> Sorry for my delay in answering over the weekend.
> And please don't remove the I18N IG mailing list address.
> 
> At 11:59 03/07/07 -0500, pat hayes wrote:
> 
> >>>  - Whereas the XML conventions for real datatypes in many 
> ways can be
> >>>     taken as just a notational convention for abstract 
> concepts such as
> >>>     'integer' that RDF treats as abstract concepts, in the case of
> >>>     XML literals, we are dealing with marked-up text, and 
> so there the
> >>>     abstraction we are dealing with is XML, not just the notation.
> >>>     (if RDF would want to create their own abstraction of 
> marked-up
> >>>     text, that would be a different thing, but currently, 
> it doesn't)
> >>
> >>Again, you seem to be presuming that if it is an XML literal, it
> >>is natural language content. That presumption unfounded.
> >
> >In fact, the very existence of RDF/XML illustrates this. 
> Like it or not, 
> >RDF/XML is legal XML, so can itself be enclosed in an RDF 
> XML literal; but 
> >one would not expect that RDF/XML to inherit any attributes 
> of the outer 
> >RDF/XML.
> 
> Yes, you can. But that's not the primary goal of XML literals, 

Er. That might not be *your* primary goal...

But that's why it is crucial to maintain the division
between the encapsulated literal and the encapsulating
markup -- because given the broad applicability of XML,
we cannot *know* what the goal of those literals are,
and thus, we need to be sure we don't stomp all over
someone's application space by infecting those *literals*
with contextual artifacts.

> and
> that's not what they are usually used for. 

Again, I know *alot* of folks who will disagree with 
you there, myself included.

> Let's not design things
> so that we can make a point, but so that they are most useful for
> what they are most used for.

Rather, let's choose a design that is general, and agnostic to how 
XML literals are used, and equally supportive of all RDF applications.

> And by the way, coming back to one of the main points, plain literals
> do inherit language information from the context (if there is such
> information), and there is always xml:lang="" 

But as has been pointed out, it should not even be possible,
using the machinery of the encapsulating markup, to modify
the semantics of the encapsulated content.

> if that's not desirable,
> and on the other hand, there is no guarantee that plain literals are
> natural text.

I agree. And I've already noted that the infection of plain
literals by xml:lang is a bug, but one that legacy considerations
prevent us from fixing this time around.

Patrick

--
Patrick Stickler
Nokia, Finland
patrick.stickler@nokia.com

Received on Wednesday, 9 July 2003 05:53:43 UTC