Re: [External] : labelled property graphs vs -star extension of RDFn vs -star extension of named graphs

The optimization for storing and querying the data is a vendor-specific issue. I doubt if users even care as to how hard it is for the vendors to optimize. (Vendors are on their own for that.) Instead, the problem here is that the data creator has to enter that many extra triples or have a non-uniform structure of the data and the query designer has to specify a more complex queries or have to rewrite some of the queries.

Thanks,
Souri.

________________________________
From: Peter F. Patel-Schneider <pfpschneider@gmail.com>
Sent: Thursday, December 14, 2023 11:50 AM
To: Souripriya Das <souripriya.das@oracle.com>; RDF-star Working Group <public-rdf-star-wg@w3.org>
Subject: Re: [External] : labelled property graphs vs -star extension of RDFn vs -star extension of named graphs

I find it misleading to just count triples and claim superiority over fewer
triples when the triples are different between different formalisms.   One has
to consider how complex the triples are and how they can be stored and
accessed in an optimized system.

peter