Re: Definition of RDF Dataset Merge added to SPARQL query draft

Hmm.

I don't think that this definition is well defined.

Let m(Ga,Gb) be the merge of Ga and Gb

Then, from the definition, the merge of
    DS1 = { G1, (<u11>, G11), (<u12>, G12), ... [,] (<u1n>, G1n) }
and DS2 = { G2, (<u21>, G21), (<u22>, G22), ... [,] (<u2m>, G2m) }
could be DS = { m(G1,G2) }

There are other "solution"s for the merge.

There are lots of other minor issues in the "definition" as well.

You probably should also be clear in defining "distinctness" for URIs.
It appears that there has been considerable discussion on this topic.

peter

PS:  By the way, it might be a good idea not to use "default graph" in
the context of an RDF Dataset.


From: Lee Feigenbaum <lee@thefigtrees.net>
Subject: Definition of RDF Dataset Merge added to SPARQL query draft
Date: Tue, 5 Apr 2011 12:26:27 -0500

> [RDF WG member hat off, SPARQL WG chair hat on]
> 
> The SPARQL WG has added a definition of the merge of two RDF Datasets to 
> the SPARQL Query editors' draft. The definition is at:
> 
> http://www.w3.org/2009/sparql/docs/query-1.1/rq25.xml#defn_RDFDatasetMerge
> 
> The hope is that if the RDF WG should choose to base graph work on the 
> concept of an RDF Dataset that this merge definition will be useful for 
> some of that work. (See, for instance, the thread that followed from 
> Richard's proposal in 
> http://lists.w3.org/Archives/Public/public-rdf-wg/2011Mar/0277.html .)
> 
> The SPARQL WG is hoping to advance documents to Last Call by the end of 
> this month, so any feedback on this definition would be appreciated 
> soon. Perhaps this is a topic that can be discussed at the upcoming F2F?
> 
> thanks,
> Lee
> On behalf of the SPARQL WG
>

Received on Tuesday, 5 April 2011 17:59:54 UTC