- From: William Waites <ww@styx.org>
- Date: Tue, 4 Jan 2011 13:47:25 +0100
- To: Michael Brunnbauer <brunni@netestate.de>
- Cc: semantic-web@w3.org
* [2011-01-03 18:45:59 +0100] Michael Brunnbauer <brunni@netestate.de> écrit: ] For a dataset of this size, we could make use of a dump in nquads format. Ok, I've been buried under some other wokr but will hopefully get a dump made soon. ] Would it be possible to use less blank nodes ? A construct like ] ] <rdf:Description rdf:nodeID="b11127987"> ] <skos:notation>Black Swan</skos:notation> ] <owl:sameAs rdf:resource="http://bibliographica.org/entity/3683b388bab11b8e411049f7774ee2b7"/> ] ... ] </rdf:Description> ] ] in http://bnb.bibliographica.org/entry/GB98Z7613.rdf seems unnecessary to me. There's a method to this madness. The BL data comes as all blank nodes, there's not a single URI to be found. So what we wanted to do was to change it as little as possible but still be able to add annotations. This is the reason for the owl:sameAs that is just adding one triple instead of removing several and adding several new ones. This is also related to deduplication of authors and publishers which is not an easy problem. Cheers, -w -- William Waites <mailto:ww@styx.org> http://eris.okfn.org/ww/ <sip:ww@styx.org> 9C7E F636 52F6 1004 E40A E565 98E3 BBF3 8320 7664
Received on Tuesday, 4 January 2011 12:48:28 UTC