To sum up: * Evgeny seems to be willing to stand in as temporary maintainer - if no one ever contributes, we kill this project after a few months (ok?) * The basic approach is accepted and a basis for future improvements - lets release it. Based on more use cases, we will develop it * there is no need to store the data in the triplestore, BUT describing it in an ontology may help communicating between some people and servcies. and after all, RDF is also XML, so it may end up somewhere in the code after all
ok? merry christmas, best Leo It was Evgeny Egorochkin who said at the right time 14.12.2009 18:42 the following words: > В сообщении от Пятница 11 декабря 2009 15:14:14 автор Ivan Frade написал: > >> Hi, >> >> On Fri, Dec 11, 2009 at 3:02 PM, Roberto Guido <[email protected]> wrote: >> >>> On Fri, 2009-12-11 at 13:47 +0100, Leo Sauermann wrote: >>> >>>> this is going to be a great ontology that really can enable nice >>>> applications to happen! >>>> >>> Probably people from Gnome Zeitgeist are also interested, I will try to >>> contact them. >>> >> I am indirectly involved in zeitgeist (wrote a couple of lines of code and >> used it in a prototype), i was in the hackfest and i can talk about our >> point of view about how to integrate tracker and zeitgeist (i.e. triplet >> store and activity logging) >> >> Our solution is Zeitgeist storing all the history, running algorithms and >> spitting refined data in the triplet store. In other words: Tracker is a >> snapshot of the information, and Zeitgeist is the journal of how the user >> got there. >> >> Storing all the history of activities of the user will spam any triplet >> store (it doesn't matter how efficient it is), and doesn't provide any >> direct benefit, because the value is in data-mining/analyze that raw data. >> Basically this means that we don't need an action ontology in the store >> (maybe it is useful for Zeitgeist itself, considering ontology as data >> model... but i doubt a triplet store should be used at all there). >> > > I wouldn't go as far as to say it will spam the triple store. Say if you add > a > triplet like <resource> nuao:used "SomeDate", you're going to accumulate > maybe > 100 triples over the lifetime of the resource as compared to say 30 triples > describing the resource. While this is a lot, I don't see this completely > killing the backend. > > You are correct in that we still need to store aggregates and this ontology > is > the place for this. > > One more possibility is storing aggregates and recent activity. We definitely > need a list of use cases. > > -- _____________________________________________________ Dr. Leo Sauermann http://www.dfki.de/~sauermann Deutsches Forschungszentrum fuer Kuenstliche Intelligenz DFKI GmbH Trippstadter Strasse 122 P.O. Box 2080 Fon: +43 6991 gnowsis D-67663 Kaiserslautern Fax: +49 631 20575-102 Germany Mail: [email protected] Geschaeftsfuehrung: Prof.Dr.Dr.h.c.mult. Wolfgang Wahlster (Vorsitzender) Dr. Walter Olthoff Vorsitzender des Aufsichtsrats: Prof. Dr. h.c. Hans A. Aukes Amtsgericht Kaiserslautern, HRB 2313 _____________________________________________________
_______________________________________________ Xesam mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/xesam
