W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > July to September 2010

Re: SPARQL Update - meta data?

From: Andy Seaborne <andy.seaborne@epimorphics.com>
Date: Tue, 14 Sep 2010 14:23:12 +0100
Message-ID: <4C8F7740.2040900@epimorphics.com>
To: Lee Feigenbaum <lee@thefigtrees.net>
CC: SPARQL Working Group <public-rdf-dawg@w3.org>


On 14/09/10 14:15, Lee Feigenbaum wrote:
> On 9/14/2010 5:11 AM, Andy Seaborne wrote:
>> From work in the UK Government Linked Data Project, there is discussion
>> of sending around changes to graphs. Part of this is to include
>> information to describe the change.
>>
>> One way is to use TriG files as bundles of graphs. Another is to use
>> SPARQL Update but SPARQL Update is missing the ability to include meta
>> data about the update request, or even link to meta about the request.
>>
>> Comments have traditionally been one way to do this but having a
>> standard machine accessible way would be useful.
>>
>> Proposal:
>>
>> Add to the SPARQL Update format:
>>
>> META { ... }
>>
>> where an operation can go. All the META in a request are accumulated
>> into a single graph (or dataset?). This RDF is just carried with the
>> request and does not affect the execution.
>>
>> The contents of the META are grammar rule "TriplesBlock" without
>> variables (or "Quads" without variables if it's a dataset.)
>>
>> Linking could be as simple as
>>
>> META { <> rdfs:seeAlso <http://example/other> }
>>
>> The choice of the word META was just taken from the style of HTML and
>> RDFa.
>
> This is not totally unlike:
> http://www.w3.org/2009/sparql/wiki/Feature:Pragmas right?

Not really - pragmas, as I understand them are an extension mechanism 
that can be used to modify the execution in some way such as setting 
system variables or declaring an extension.

Adding metadata about the request (author, description, date) is 
specifically not changing the execution in anyway - just bundling some 
data to be carried with it.

	Andy

> I'm hesitant to add something new at this point, though I think it is
> useful.
>
> Lee
>
>
>> Andy
>
Received on Tuesday, 14 September 2010 13:23:54 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:15:43 GMT