- From: Steve Harris <steve.harris@garlik.com>
- Date: Tue, 24 May 2011 13:10:49 +0100
- To: Richard Cyganiak <richard@cyganiak.de>
- Cc: Andy Seaborne <andy.seaborne@epimorphics.com>, Pat Hayes <phayes@ihmc.us>, public-rdf-wg@w3.org
On 2011-05-24, at 12:09, Richard Cyganiak wrote:
> On 24 May 2011, at 10:39, Steve Harris wrote:
>> The difference would/could be in DATATYPE("foo"@en) in SPARQL.
>
> Yes.
>
>> If people write "foo"^^lang:en, then there will be some issues with pre-2011 systems.
>
> ??
>
> I don't understand. What is lang:en? The proposal does not allow any new syntax, and in fact states: “7. In concrete syntaxes, the "foo"@en form MUST be used for rdf:LanguageTaggedStrings.” [1]
>
>
>> The things I'm keen to avoid are e.g. in SPARQL Results, not having to emit:
>>
>> <binding><literal datatype="http://www.w3.org/2001/XMLSchema#string">foo</literal></binding>
>>
>> For each plain literal result.
>
> I have trouble parsing the double negation ...
>
> The intention of the proposal is that *both* of the following would still be valid:
>
> <binding><literal>foo</literal></binding>
> <binding><literal datatype="&xsd;string">foo</literal></binding>
>
> But implementations SHOULD use the first: “In concrete syntaxes, the "foo" form SHOULD be used instead of "foo"^^xsd:string” [1]
Sorry, I must have got two threads confused.
I agree with Andy, this is not going to play well with existing systems, unless rdf:LanguageTaggedString exists purely in the abstract syntax, you can still end up with literals with both a datatype, and a language tag.
- Steve
--
Steve Harris, CTO, Garlik Limited
1-3 Halford Road, Richmond, TW10 6AW, UK
+44 20 8439 8203 http://www.garlik.com/
Registered in England and Wales 535 7233 VAT # 849 0517 11
Registered office: Thames House, Portsmouth Road, Esher, Surrey, KT10 9AD
Received on Tuesday, 24 May 2011 12:11:19 UTC