- From: Andy Seaborne <andy.seaborne@epimorphics.com>
- Date: Wed, 14 Nov 2012 20:47:20 +0000
- To: Pat Hayes <phayes@ihmc.us>
- CC: public-rdf-wg@w3.org
On 14/11/12 17:16, Pat Hayes wrote: >>> [[ There is a widely held misconception in the RDF community that >>> graphs cannot share blank nodes. This stems from the fact in >>> practice they rarely do and almost never need to, ]] >>> >>> Any systems that has the default query graph as the union of >>> named graphs is sharing bnodes across graphs. >>> >>> Such systems are not 'rare'. > But if you take two graphs from random places on the Web, do they > share bnodes? Most people think they obviously don't, but the specs > implicitly say they might (by conspicuously not saying they don't.) Agree - what really keeps the bNodes apart is that they deserialized/serialized as part of the transport process. Their event horizon. Until skolemization ... which reveals the singularity inside the event horizon. >> 2/ Talk about "fresh blank node" and not the universal not-quite-arbitrary set of blank nodes. > > The trouble for me with this way of talking is that it seems to suggest some kind of global source for new, fresh bnodes, like the garden of Eden being the single source of the four rivers. But where is this magic bnode spring? I guess maybe TImBL is the only person who knows? FWIW talking about "fresh blank nodes" has never seemed to cause confusion elsewhere. Andy
Received on Wednesday, 14 November 2012 20:47:56 UTC