[dsdl-discuss] Re: What is the best way to simulate

Rick

Re:
>The technical answer is that that restricting anyType with mixed content
and a null sequence is a typical way that people get text content in XML
Schemas without typing that content: ugly isnt it? XSD makes it easier to
say data content has a type (like xs:string) rather than saying data
content has no type. Actually this declaration prevents any further
datatyping of the DC data content, because of mixed='true'.

Wouldn't it be simpler to make the model for the DC elements just "text"? Do
you really mean to suggest that users can embed other elements within DC
element contents to create mixed content? If so, how do you get to convert
it to RDF?

Martin

----- Original Message -----
From: "Rick Jelliffe" <rjelliffe@allette.com.au>
To: <dsdl-discuss@dsdl.org>
Cc: <dsdl-discuss@dsdl.org>
Sent: Friday, August 18, 2006 5:35 PM
Subject: [dsdl-discuss] Re: What is the best way to simulate

Martin Bryan said:
> Rick
>
> I'm confused (not an unusual situation I know!). Why is SimpleLiteral
> based
> on anyType rather than being a simple string? Are you expecting someone to
> want to extend these definitions by specifying a specific type, such as
> xsd:date for the date element? What other DC elements would need typing?
>

The real answer is this: the XSD schemas are supplied by MS as their
work-in-progress for MSOOX. I can only control the RELAX NG output, and
have to accept whatever input anyone provides. However, at least in this
case, the RELAX NG output is clearer than the XSD input, I think(?)

The technical answer is that that restricting anyType with mixed content
and a null sequence is a typical way that people get text content in XML
Schemas without typing that content: ugly isnt it? XSD makes it easier to
say data content has a type (like xs:string) rather than saying data
content has no type. Actually this declaration prevents any further
datatyping of the DC data content, because of mixed='true'.

And yes, Martin, you are correct that other schema modules define other
metadata elements in terms of these Dublin Core elements. They use
substitution groups to express some semantic similarity. So that a
type-aware query can retrieve elements by their DC equivalent no matter
what name is used. I think a processor could be made using DSRL or the
RELAX NG combination; it is just establishing the DC elements as
architectural forms really.

Cheers
Rick Jelliffe

--
DSDL members discussion list
To unsubscribe, please send a message with the
command  "unsubscribe" to dsdl-discuss-request@dsdl.org
(mailto:dsdl-discuss-request@dsdl.org?Subject=unsubscribe)
--
DSDL members discussion list
To unsubscribe, please send a message with the
command  "unsubscribe" to dsdl-discuss-request@dsdl.org
(mailto:dsdl-discuss-request@dsdl.org?Subject=unsubscribe)