On Tue, Aug 16, 2011 at 3:37 PM, Saašha Metsärantala <[log in to unmask]> wrote:
> Hello!
>
>> if EDTF is a success, it will be used outside of XML.
>
> I hope it will both!
>
>> > we may want to use QNames, URL, URI, IRI,
>> > LEIRI, URN ...
>> we should avoid any solutions that are too
>> particularly tied to XML (like QNames).
>
> I wrote that we "may want", but maybe I should have clarified that we may
> choose something else, of course! That's a question for phase two.
>
> The datatype xs:QName is defined in the context of XML Schema - not XML per
> se. Furthermore, it is usable in non-XML contexts. The W3C explains:
>
>> The value space of QName is the set of tuples
>> {namespace name, local part}, where namespace
>> name is an anyURI and local part is an NCName.
>
> http://www.w3.org/TR/xmlschema11-2/#NCName
>
> To use xs:QName within JSON, you will need to write this tuple using a
> notation compatible with JSON, for example:
>
> { localPart : "something", namespaceName : "http://www.example.com/NS" }
>
> with a one-to-one matching to the value space described by the W3C (assuming
> suitable escaping).
I know. But QNames and other similar prefix-based indirection are
highly-controversial. See recent debates around microdata vs. rdfa.
I don't have a problem with myself though.
Bruce
