Re: Proposed response to DB-1

The rewording you had is fine.

On 19/10/10 16:07, Alexandre Passant wrote:
>
> On 19 Oct 2010, at 14:54, Andy Seaborne wrote:
>
>> Looks good to me.
>>
>> FWIW:
>>
>> The URI is naming a document which is the results of the CONSTRUCT query, which is sent as a representation over the wire from which the triples are parsed out - I don't think the doc needs changing. CONSTRUCT isn't special in anyway (e.g. DESCRIBE).
>>
>
> Right, nothing special but seems (based on the email), that it requires clarification ... to say that it's not special.
>
>> So
>>
>> <documentURI>  =>  <url>  or dereferencable URI
>
> Ok, so
>
> "The documentURI specifies the URI of a document or the URL of a service such that a store will be able to identify, locate and read the document's content".

<Spec mode>
it's not the URI of a service - it's the service and query string.  The 
URL really does name a document (the results), it's just a computed one.
</Spec mode>

GET of a service URL returns the service description which is might be a 
bit surprising as the result of a CONSTRUCT query.

	Andy

>
> Makes sense ?
>
> Alex.
>

Received on Tuesday, 19 October 2010 15:12:55 UTC