On Tue, Feb 3, 2015 at 4:44 PM, Andy Seaborne <a...@apache.org> wrote:
> > > It depends a bit on the "A" in API -- whether "application" means "user > application" or includes system-machinery things like parsers but the > boundary is nit easy to fix. > I don't see the reason to make such a distinction. What's the difference between a piece of software that converts a file to RDF and one that describes the music it hears in RDF, with or without involving a human. The relevant question is if the API should only allow for the same to work on data provided by different implementations (so that one can e.g. switch from jena to sesame) or if it should allow to mix data from different implementations. Clerezza does the latter, you can have a weather service proving a graph (not stored anywhere but computed just in time from the data of the sensors) and a triple store copy data from one graph to the other (well, in this case only in one direction, as currently the weather is read only). Cheers, Reto
