Hi Kingsley & all, Wikidata RDF URIs are of the form http://www.wikidata.org/entity/Q42 , not http://www.wikidata.org/entity/Q42#this . Maybe the response codes of Wikidata's HTTP server are not quite correct yet, but http://www.wikidata.org/entity/Q42 is not meant to be the URI of a document about Douglas Adams. It's meant to be the Wikidata RDF URI for the entity otherwise known as Douglas Adams.
I also didn't understand how appending #this helps with our current problem. As I understand it, the problem is that DBpedia wants to publish data about Wikidata items, but that data won't be returned by HTTP requests to Wikidata URIs. Regards, Christopher On 31 August 2013 19:58, Kingsley Idehen <kide...@openlinksw.com> wrote: > On 8/31/13 8:08 AM, Sebastian Hellmann wrote: > > Hi everybody, > we should focus on features, we can provide: > > Major Feature 1: > loading data for SPARQL queries. This is independent from the URIs and we > might use the Wikidata URIs or new ones. > > > Wikidata URIs as per the mail I just sent in response to Dimitris. > > > Major Feature 2: > providing linked data for the data in the store. > If we reuse Wikidata URIs, we will end up having data in the store, which is > not crawlable via linked data. > > > The "#this" heuristic gets you around that problem. Thus, the > follow-your-nose pathways remain intact, modulo any bugs in the actual RDF > published by Wikidata (e.g., there are some bugs right now re. datatypes > e.g., xsd:integer missing an "r" so you have xsd:intege etc.. Just typo > style bugs in some places). > > > I see this as an essential feature, since we provide for now: > 1. flat, easy RDF > 2. properties mapped to the DBpedia Ontology > 3. wikidata entities typed via the DBpedia Ontology > 4. improved datatypes refined for compatability with xsd > > > Yes. > > > Suggesting the use of Wikidata URIs will disallow retrieving: > <http://www.wikidata.org/entity/QXXX> a dbo:Person . > <http://www.wikidata.org/entity/QXXX> dbo:birthPlace "1960-01-01"^^xsd:date > . > > > See my example: > > > ## Wikidata RDF document describes a 'Person' class > <http://www.wikidata.org/entity/Q215627> > <http://open.vocab.org/terms/describes> > <http://www.wikidata.org/entity/Q215627#this> . > > ## Wikidata 'Person' class is equivalent to DBpedia 'Person' class > <http://www.wikidata.org/entity/Q215627#this> > owl:equivalentClass dbpedia-owl:Person . > > > ABox (Assertions): > > ## Wikidata RDF Document describes 'Douglas Adams' but without explicit > denotation via a URI > ## thus, "#this" heuristic is used to derive a Linked Data URI for 'Douglas > Adams' from the Wikidata RDF Document URL. > > <http://www.wikidata.org/entity/Q42> > <http://open.vocab.org/terms/describes> > <http://www.wikidata.org/entity/Q42#this> ; > foaf:primaryTopic <http://www.wikidata.org/entity/Q42#this> . > > ## derived URI for 'Douglas Adams' is then used as the subject or object of > an owl:sameAs relation with the DBpedia URI for 'Douglas Adams' > > <http://www.wikidata.org/entity/Q42#this> > owl:sameAs <http://dbpedia.org/resource/Douglas_Adams> . > > > > while this will do: > <http://wd.dbpedia.org/entity/QXXX> a dbo:Person . > <http://wd.dbpedia.org/entity/QXXX> dbo:birthPlace "1960-01-01"^^xsd:date . > <http://wd.dbpedia.org/entity/QXXX> owl:sameAs > <http://www.wikidata.org/entity/QXXX> . > > > You don't need that, if you follow the pattern outlined in the example > above. You arrive at the very same place via: > > > <http://www.wikidata.org/entity/Q42#this> owl:sameAs > <http://dbpedia.org/resource/Douglas_Adams> . > > Both URIs resolve, and if owl:sameAs reasoning is applied, you get the union > based mesh (smush) of the triples describing both of the entities > participating in the aforementioned relation. > > > > I am against losing this feature. Please everybody look at: > https://www.wikidata.org/wiki/Special:EntityData/Q42.ttl > You can find *only* labels and language links, not the "cool" data. > > > Yes, because that's a Turtle document, as per my example. > > Furthermore, I doubt that this will be available any time soon. Cleaning > datatypes is no easy process. Hady is still young, but already quite a good > and goal-oriented programmer and working overtime, so it is quite an > investment, which is probably not high on their list. > > Failing to provide crawlable linked data will likely result in an effort > like dbpedialite to fill the gap. > We might shorten wikidata.dbpedia.org to wd.dbpedia.org though or just > data.dbpedia.org > > @Kingsley and JC: do you agree? > > > I agree with the fact that any mapping endeavor MUST keep follow-your-nose > pattern crawling intact. I also believe we can achieve this via "#this" > heuristic which ultimately save time and work in the future, one in which > Wikidata will get better at publishing RDF based documents that bear Linked > Data :-) > > > Kingsley > > > -- > Sebastian > > > Am 31.08.2013 09:20, schrieb Dimitris Kontokostas: > > Hi Kingsley, > > Sorry but I still don't see a clear solution with your suggestion. > Just like in Wikipedia when we see an Infobox_Person we assume that the > resource is an dbo:Person > we want to do the same in WikiData when we see a P107 Q215627 claim > In the 1st case it's an unstructured text to LD the 2nd is an LD2LD but to a > less complicated and well-known schema (that will make existing DBpedia > queries compatible to Wikidata) > > Thus, When I wrote the triple Q42 a dbo:Person, that triple did not came > from dbr:Douglas_Adams as you replied but directly from the Wikidata claim > > Now, does Q42 need to be the original Wikidata resource > <http://www.wikidata.org/entity/Q42> and if yes, how should DBpedia publish > & dereference these triples? > Or should we use our own domain and make owl:sameAs / foaf:primaryTopic > links to Wikidata like we did with all Wikipedia language editions? > > Other suggestions / directions are also welcome of course! > > Best, > Dimitris > > > > > On Fri, Aug 30, 2013 at 7:45 PM, Kingsley Idehen <kide...@openlinksw.com> > wrote: >> >> On 8/30/13 12:25 PM, Dimitris Kontokostas wrote: >> >> Thanks Kingsley, >> >> for now we don't host the data so wikidata.dbpedia.org/resource/.. does >> not dereference anything >> >> lets take an example to recup >> Q42 (Douglas Adams), besides the answer to the ultimate question [1] :) >> has the following triple >> http://www.wikidata.org/entity/Q42 http://www.wikidata.org/entity/P107 >> http://www.wikidata.org/entity/Q215627 >> >> which in our dump *will* be translated to >> http://wikidata.dbpedia.org/resource/Q42 owl:sameAs >> http://www.wikidata.org/entity/Q42 >> http://wikidata.dbpedia.org/resource/Q42 rdf:type dbpedia-owl:Person >> >> If I get it right, your suggestion is to generate this triple directly >> right? >> http://www.wikidata.org/entity/Q42 rdf:type dbpedia-owl:Person >> >> but in this case, when someone opens Q42 he/she will not see out added >> triples >> >> >> Having taken a closer look at the RDF content in WikiData documents It >> should be: >> >> <http://www.wikidata.org/entity/Q42> foaf:primartyTopic >> <http://dbpedia.org/resource/Douglas_Adams> . >> >> To be really safe, i.e., not vulnerable to changes relating to how they >> evolve handling of entity disambiguation via content negotiation etc., you >> can opt to be much more specific about the RDF document URLs: >> >> >> <http://www.wikidata.org/entity/Q42.nt> foaf:primaryTopic >> <http://dbpedia.org/resource/Douglas_Adams> . >> http://www.wikidata.org/entity/Q42.rdf> foaf:primaryTopic >> <http://dbpedia.org/resource/Douglas_Adams> . >> >> That's it, let the Web of Linked Data do the rest :-) >> >> >> Kingsley. >> >> >> >> Cheers, >> Dimitris >> >> >> [1] >> http://en.wikipedia.org/wiki/42_%28number%29#The_Hitchhiker.27s_Guide_to_the_Galaxy >> >> >> On Fri, Aug 30, 2013 at 5:57 PM, Kingsley Idehen <kide...@openlinksw.com> >> wrote: >>> >>> On 8/29/13 2:45 PM, Hady elsahar wrote: >>> >>> >>> i guess what we need now is to map between the original WikiData URIs and >>> their URIs we use in DBpedia under the new namespace , for example : >>> >>> <http://wikidata.dbpedia.org/resource/Q1293045> owl:sameAs >>> <http://wikidata.org/entity/Q1293045> >>> >>> No. >>> >>> As per my last two mails. You are making unnecessary work here. Wikidata >>> is publishing RDF based Linked Data. Thus, simply use their URIs. >>> >>> Example of the implications of your current approach: >>> >>> curl -ILH "Accept: text/turtle" >>> http://wikidata.dbpedia.org/resource/Q1293045 >>> curl: (6) Couldn't resolve host 'wikidata.dbpedia.org' >>> >>> >>> Just map the WikiData URI to the DBpedia URI and you leverage the >>> benefits of: >>> >>> >>> curl -ILH "Accept: text/turtle" http://wikidata.org/entity/Q1293045 >>> HTTP/1.0 301 Moved Permanently >>> Date: Fri, 30 Aug 2013 14:51:49 GMT >>> Server: Apache >>> Location: http://www.wikidata.org/entity/Q1293045 >>> Content-Type: text/html; charset=iso-8859-1 >>> >>> >>> HTTP/1.0 302 Moved Temporarily >>> Date: Fri, 30 Aug 2013 14:51:49 GMT >>> Server: Apache >>> Location: https://www.wikidata.org/wiki/Special:EntityData/Q1293045 >>> Content-Type: text/html; charset=iso-8859-1 >>> >>> >>> HTTP/1.1 303 See Other >>> Server: nginx/1.1.19 >>> Date: Fri, 30 Aug 2013 14:51:54 GMT >>> Content-Type: text/html; charset=utf-8 >>> Connection: close >>> X-Content-Type-Options: nosniff >>> Cache-Control: private, s-maxage=0, max-age=0, must-revalidate >>> Vary: Accept-Encoding,X-Forwarded-Proto,Cookie >>> Expires: Thu, 01 Jan 1970 00:00:00 GMT >>> Last-Modified: Fri, 30 Aug 2013 14:51:54 GMT >>> Location: https://www.wikidata.org/wiki/Special:EntityData/Q1293045.ttl >>> >>> >>> HTTP/1.1 200 OK >>> Server: nginx/1.1.19 >>> Date: Fri, 30 Aug 2013 14:51:55 GMT >>> Content-Type: text/turtle; charset=UTF-8 >>> Content-Length: 8388 >>> >>> -- >>> >>> Regards, >>> >>> Kingsley Idehen >>> Founder & CEO >>> OpenLink Software >>> Company Web: http://www.openlinksw.com >>> Personal Weblog: http://www.openlinksw.com/blog/~kidehen >>> Twitter/Identi.ca handle: @kidehen >>> Google+ Profile: https://plus.google.com/112399767740508618350/about >>> LinkedIn Profile: http://www.linkedin.com/in/kidehen >>> >>> >>> >>> >>> >>> >>> ------------------------------------------------------------------------------ >>> Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more! >>> Discover the easy way to master current and previous Microsoft >>> technologies >>> and advance your career. Get an incredible 1,500+ hours of step-by-step >>> tutorial videos with LearnDevNow. Subscribe today and save! >>> >>> http://pubads.g.doubleclick.net/gampad/clk?id=58040911&iu=/4140/ostg.clktrk >>> _______________________________________________ >>> Dbpedia-discussion mailing list >>> Dbpedia-discussion@lists.sourceforge.net >>> https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion >>> >> >> >> >> -- >> Kontokostas Dimitris >> >> >> >> -- >> >> Regards, >> >> Kingsley Idehen >> Founder & CEO >> OpenLink Software >> Company Web: http://www.openlinksw.com >> Personal Weblog: http://www.openlinksw.com/blog/~kidehen >> Twitter/Identi.ca handle: @kidehen >> Google+ Profile: https://plus.google.com/112399767740508618350/about >> LinkedIn Profile: http://www.linkedin.com/in/kidehen >> >> >> >> > > > > -- > Kontokostas Dimitris > > > ------------------------------------------------------------------------------ > Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more! > Discover the easy way to master current and previous Microsoft technologies > and advance your career. Get an incredible 1,500+ hours of step-by-step > tutorial videos with LearnDevNow. Subscribe today and save! > http://pubads.g.doubleclick.net/gampad/clk?id=58040911&iu=/4140/ostg.clktrk > > > > _______________________________________________ > Dbpedia-discussion mailing list > Dbpedia-discussion@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion > > > > -- > Dipl. Inf. Sebastian Hellmann > Department of Computer Science, University of Leipzig > Events: > * NLP & DBpedia 2013 (http://nlp-dbpedia2013.blogs.aksw.org, Extended > Deadline: *July 18th*) > * LSWT 23/24 Sept, 2013 in Leipzig (http://aksw.org/lswt) > Venha para a Alemanha como PhD: http://bis.informatik.uni-leipzig.de/csf > Projects: http://nlp2rdf.org , http://linguistics.okfn.org , > http://dbpedia.org/Wiktionary , http://dbpedia.org > Homepage: http://bis.informatik.uni-leipzig.de/SebastianHellmann > Research Group: http://aksw.org > > > > -- > > Regards, > > Kingsley Idehen > Founder & CEO > OpenLink Software > Company Web: http://www.openlinksw.com > Personal Weblog: http://www.openlinksw.com/blog/~kidehen > Twitter/Identi.ca handle: @kidehen > Google+ Profile: https://plus.google.com/112399767740508618350/about > LinkedIn Profile: http://www.linkedin.com/in/kidehen > > > > > > ------------------------------------------------------------------------------ > Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more! > Discover the easy way to master current and previous Microsoft technologies > and advance your career. Get an incredible 1,500+ hours of step-by-step > tutorial videos with LearnDevNow. Subscribe today and save! > http://pubads.g.doubleclick.net/gampad/clk?id=58040911&iu=/4140/ostg.clktrk > _______________________________________________ > Dbpedia-discussion mailing list > Dbpedia-discussion@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion > ------------------------------------------------------------------------------ Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more! Discover the easy way to master current and previous Microsoft technologies and advance your career. Get an incredible 1,500+ hours of step-by-step tutorial videos with LearnDevNow. Subscribe today and save! http://pubads.g.doubleclick.net/gampad/clk?id=58040911&iu=/4140/ostg.clktrk _______________________________________________ Dbpedia-discussion mailing list Dbpedia-discussion@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion
