thanks, you comments are very helpful
________________________________ Von: Joshua TAYLOR <[email protected]> An: Yana Panchenko <[email protected]> CC: "[email protected]" <[email protected]> Gesendet: 17:48 Freitag, 20.Juni 2014 Betreff: Re: nested named graphs On Fri, Jun 20, 2014 at 11:29 AM, Yana Panchenko <[email protected]> wrote: > Yes, that's page I found the example. There, G2 is a graph that contains > triples describing the signature of G1 and G1 itself as nested graph (the > latter you omitted from the example in your reply). It not clear to me how > should I represent such data (when I want G1 to include triples from G2 > preserving the information that these triples constitute G2) so that it > coforms to the current RDF grammar and is supported by Jena. I'm not aware that RDF datasets *have* any concept of nested graphs, though. I'm also not at all clear why this example needs it. There's a difference between X == "Manu Sporny is a person" S2 says that 'S1 says X.' and S2 says that 'X == "Manu Sporny is a person"' S2 says that 'S1 says X'. The not-quite-Trig on that page (from 2011, by the way) is like the second example here, whereas my updated version is like the first. The question then is, do we actually know what the content X is, or do we only know what S2 says it is? Since there's not a standard way of reifying "graph g { s p o }", I don't think that there's a standard way of saying "S says that G contains the triple T", so for the moment it's better to say "G1 contains the triple T. G2 contains the triple 'S1 affirms G1'. S2 affirms G2." -- Joshua Taylor, http://www.cs.rpi.edu/~tayloj/
