- From: Orri Erling <erling@xs4all.nl>
- Date: Tue, 10 Mar 2009 09:03:59 +0100
- To: "'Seaborne, Andy'" <andy.seaborne@hp.com>, "'Lee Feigenbaum'" <lee@thefigtrees.net>, "'SPARQL Working Group'" <public-rdf-dawg@w3.org>
All I would suggest that we take, as an informal goal statement of this iteration of the SPARQL process to make SPARQL at least capable of expressing the TPC H workload. If we are to make the web into a database, then this database should at the minimum provide the level of querying that people have come to take for granted in the relational world. Further, in the context of mapping SPARQL to SQL for on the fly integration of data sources, the proposition is sort of a non starter if the expressivity that any business intelligence person would take as the minimum to be expected is not present or depends on vendor-specific extensions. In terms of the proposed features, this goal would be met by inclusion of projection expressions, subselects, negation, aggregation and grouping. Regards Orri
Received on Tuesday, 10 March 2009 08:10:55 UTC