- From: Michael Schneider <schneid@fzi.de>
- Date: Wed, 3 Oct 2007 11:49:26 +0200
- To: "Pat Hayes" <phayes@ihmc.us>
- Cc: "Danny Ayers" <danny.ayers@gmail.com>, "Owl Dev" <public-owl-dev@w3.org>, "Dan Brickley" <danbri@danbri.org>, "Bijan Parsia" <bparsia@cs.man.ac.uk>
Pat Hayes wrote: > OK, I'm a strict bnodes-are-existential guy, and I'll answer the questions: > >> >>Do you believe that if you start from the following graph: >> >> s p o. >> _:x p o. >> >>and delete the second triple to get: >> s p o. >> >>that you have lost anything? > >Yes, of course: you have lost a triple. Do they MEAN THE SAME? >(Different question) Yes, they do. Still, one has to be careful, >because if these are subgraphs of a larger graph (as they almost >certainly will be) then they might well not mean the same. For >example: > >s p o. >_:x p o. >_:x q r. > >aint the same as > >s p o. >_:x q r. > >in anyone's book. > >I think the 'pure' non-lean redundancy exhibited in your example here >hardly ever occurs in practice. Maybe leanifyers aren't much deployed >because there isn't much use for them. Another example: I want to publish raw statistical data on the SemWeb. The data is aquired by monitoring people, and how they perform on some task. The same person may perform the task several times. There are no timestamps per data tuple, because there is a single session (one day), and specific timestamps would not be very useful information. There are also no universal identifiers (URIs!) for the different data tuples, because I think it suffices to be able to distinguish the data tuples within the graph (if at all, I really just want to do statistics). Here is the actual data: [ ex:person :alice ; ex:performance 21 ] [ ex:person :bob ; ex:performance 19 ] [ ex:person :alice ; ex:performance 23 ] [ ex:person :charly ; ex:performance 19 ] [ ex:person :bob ; ex:performance 19 ] [ ex:person :charly ; ex:performance 18 ] Now, I bought this "perfectly RDF conformant" (i.e. existential semantics aware) RDF framework for storing and publishing my data. When I do a simple test, by doing a SPARQL "SELECT COUNT(*)" query on this data, I realize that my database consists of only /five/ entries. What's going on? Cheers, Michael -- Dipl.-Inform. Michael Schneider FZI Forschungszentrum Informatik Karlsruhe Abtl. Information Process Engineering (IPE) Tel : +49-721-9654-726 Fax : +49-721-9654-727 Email: Michael.Schneider@fzi.de Web : http://www.fzi.de/ipe/eng/mitarbeiter.php?id=555 FZI Forschungszentrum Informatik an der Universität Karlsruhe Haid-und-Neu-Str. 10-14, D-76131 Karlsruhe Tel.: +49-721-9654-0, Fax: +49-721-9654-959 Stiftung des bürgerlichen Rechts Az: 14-0563.1 Regierungspräsidium Karlsruhe Vorstand: Rüdiger Dillmann, Michael Flor, Jivka Ovtcharova, Rudi Studer Vorsitzender des Kuratoriums: Ministerialdirigent Günther Leßnerkraus
Received on Wednesday, 3 October 2007 09:49:39 UTC