Am 18.03.2017 um 19:03 schrieb Bob DuCharme: It's a predicate (that's what RDF calls all relationships) that expresses a direct claim (as opposed to a full statement). Direct claims are one of two ways Wikidata Statements are mapped to RDF. In the wikidata query service, each statement is represented twice - once as a full statement, and once as a direct claim. Direct claims represent a "naive projection" of wikidata to RDF: everything that is claimed (by anyone) to be true (under any circumstances) is assumed to be true. So you get triples like <wd:Q64> <wdt:P1376> <wd:Q183> meaning "Berlin - capital-of - Germany". Simple to work with, but incomplete: you also get <wd:Q64> <wdt:P1376> <wd:Q27306> ("Berlin - capital-of - Kingdom of Prussia"), without an easy way to see that one is current and the other is not. To get all the additional context information, you need to look at the full statement representation, which provides a complex structure of value, qualifiers, and source references. The full mapping will use predicates with the "wds" prefix to connect the item (subject) to the structure representing the statement with all its parts. -- Daniel Kinzler Senior Software Developer Wikimedia Deutschland Gesellschaft zur Förderung Freien Wissens e.V.

Am 18.03.2017 um 21:27 schrieb Bob DuCharme: The full representation of the statement in this case is: wds:Q64-43CCD3D6-F52E-4742-B0E3-BCA671B69D2C a wikibase:Statement, wikibase:BestRank ; wikibase:rank wikibase:PreferredRank ; ps:P1376 wd:Q183 ; prov:wasDerivedFrom wdref:ba76a7c0f885fa85b10368696ab4ac89680aa073 . wdref:ba76a7c0f885fa85b10368696ab4ac89680aa073 a wikibase:Reference ; pr:P248 wd:Q451546 ; pr:P958 "Artikel 2 (1)" . This RDF representation can be found at <https://www.wikidata.org/wiki/Special:EntityData/Q64.ttl>. Content negotiation will take you there from the canonical URI, <https://www.wikidata.org/entity/Q64.ttl> In addition to the actual value, the RDF above also give the rank, and a source reference (nameley, the re-unification treaty). This statement doesn't currently have a qualifier - it should have at least one, stating since when Berlin is the Capital of Germany. That qualifier would be represented as: wds:Q64-43CCD3D6-F52E-4742-B0E3-BCA671B69D2C pq:P580 "1990-10-03T00:00:00Z"^^xsd:dateTime ; The Statement ID, Q64$43CCD3D6-F52E-4742-B0E3-BCA671B69D2C, can be found in the HTML source of the page, encoded as a CSS class. These IDs are not exposed nicely anywhere. But usually, one would look at the RDF representation right away, or at least got from HTML to *all* the RDF. HTH -- daniel -- Daniel Kinzler Senior Software Developer Wikimedia Deutschland Gesellschaft zur Förderung Freien Wissens e.V.

Am 18.03.2017 um 22:48 schrieb Bob DuCharme: The Wikidata (or technically, Wikibase) data model is not defined in terms of RDF. Have a look at the primer <https://www.mediawiki.org/wiki/Wikibase/DataModel/Primer> and the spec <https://www.mediawiki.org/wiki/Wikibase/DataModel>. Entitites are the top level elements of Wikidata. There are currently two kinds: Items (things or concepts in the world) and Properties (attributes for describing Items and other entities). Wikibase Entities are certainly Resources in the RDF sense, but so are some of the more fine grained components of the Wikibase model, such as Statements and References. You can find the OWL file for the RDF binding of Wikibase at <http://wikiba.se/ontology>. -- Daniel Kinzler Senior Software Developer Wikimedia Deutschland Gesellschaft zur Förderung Freien Wissens e.V.

Hi! Short addition to the great explanations by Daniel - the direct claim(s) have two properties: 1. They always have the simple values (i.e. either URL or literal or bnode), even though full values may contain more information. This means direct claims sometimes are missing secondary information, such as precision, globe, units, etc., but they are simpler to work with and enough for many simpler queries. 2. They include "best ranked" statements - i.e., if the property has preferred rank statements, then only preferred ones will be in direct claims. Otherwise, regular rank ones will be. Deprecated rank claims are never in direct claims. The preferred rank usually is used to describe "current" state of affairs - e.g. current manager of a company, major of a city, population of a country, spouse, etc. - or closest to it. If you need to see which of the full statements have the same data as direct claims, look for statements with type wikibase:BestRank. This way you can have the same data but with qualifiers, references and full values. -- Stas Malyshev smalyshev(a)wikimedia.org

Am 19.03.2017 um 18:21 schrieb Bob DuCharme: The current mapping is the one described at https://www.mediawiki.org/wiki/Wikibase/Indexing/RDF_Dump_Format -- Daniel Kinzler Senior Software Developer Wikimedia Deutschland Gesellschaft zur Förderung Freien Wissens e.V.

Am 27.03.2017 um 15:13 schrieb Kingsley Idehen: It's a matter of perspective. I agree that Wikidata can be *represented* as a collection of reified RDF Statements. That's what we do for the query service. But I do not agree that this is what Wikidata *is*. RDF and the Wikibase model are quite different conceptually. But they are of equal power and thus formally equivalent: one can be represented using the other. Just because a Turing Machine is computationally equivalent to lambda calculus, that does not mean they are the same thing. Understanding one in terms of the other may be helpful in some context, and irrelevant in another. There is nothing special about the relationship between Wikibase/Wikidata and RDF; Wikibase has an RDF binding, but it is not defined in terms of RDF, its specification does not rely on RDF concepts. The Wikibase model can just as well (or perhaps more easily) be understood and represented in terms of the Topic Maps model (ISO 13250). Academically, the Wikibase model could perhaps be described as an extended model logic with reasoning rules for provenance. I think W. Stelzner explored related ideas in the 80s. Maybe one day I'll find the time to dig into this some more. -- Daniel Kinzler Principal Platform Engineer Wikimedia Deutschland Gesellschaft zur Förderung Freien Wissens e.V.

On 3/27/17 10:42 AM, Daniel Kinzler wrote: My point is that your model boils down to treating statements as "first class citizens" so to speak. If true, then it is as I described i.e., still an RDF model, but with emphasis of statement reification, which is actually a good thing. RDF doesn't exclude reification. Put differently, using reification doesn't amount to a new model different from that of RDF. I am not implying that. RDF concepts boil down to the use of sentences to describe anything, just as we do everyday in the so-called real world. I think we can just agree to disagree for now, since nothing you've stated is fundamentally contrary to my view of RDF -- as a Language for describing anything (including statements) :) -- Regards, Kingsley Idehen Founder & CEO OpenLink Software (Home Page: http://www.openlinksw.com) Weblogs (Blogs): Legacy Blog: http://www.openlinksw.com/blog/~kidehen/ Blogspot Blog: http://kidehen.blogspot.com Medium Blog: https://medium.com/@kidehen Profile Pages: Pinterest: https://www.pinterest.com/kidehen/ Quora: https://www.quora.com/profile/Kingsley-Uyi-Idehen Twitter: https://twitter.com/kidehen Google+: https://plus.google.com/+KingsleyIdehen/about LinkedIn: http://www.linkedin.com/in/kidehen Web Identities (WebID): Personal: http://kingsley.idehen.net/dataspace/person/kidehen#this : http://id.myopenlink.net/DAV/home/KingsleyUyiIdehen/Public/kingsley.ttl#this

Am 27.03.2017 um 23:48 schrieb Kingsley Idehen: Yes, that's what RDF is. My pint is: just because seomthing can be described in RDF doesn't mean it *is* RDF. As you said, RDF can describe anything. If anything that can be described with RDF *is* RDF, then everything is RDF. Then the term would be meaningless. The Wikibase model "is" an RDF model just as much as it "is" modal logic system, or any other sufficiently powerful formal language. -- Daniel Kinzler Senior Software Developer Wikimedia Deutschland Gesellschaft zur Förderung Freien Wissens e.V.