Am 27.06.2013 16:08, schrieb Hady elsahar: The triple you mention makes explicit the distinction (and relationship) between the *thing* and it's *description*. This is to avoid an "identity crisis" as described here: <http://www.ontopia.net/topicmaps/materials/identitycrisis.html>. So... * http://www.wikidata.org/entity/Q1000 is the canonical URI of the "thing" Gabon. * http://www.wikidata.org/wiki/Special:EntityData/Q1000 is the canonical URI of the *document* that *describes* Gabon. That is, the document is *about* the thing. Which is what the triple expresses. Why this distinction is needed becomes clear when thinking about the description of a book: if the description has the same URI also used for the book, and you ask for the authors, should you get the authors of the book, or the authors of the description? Only if you make the distinction between the thing and it's description can you avoid this issue. Now, when using the above URIs as URLs and retrieving the respective document, you will get the same data (we can hardly return the "thing" Gabon from that URL, so we rather return the document about the thing). That does not mean that the URIs mean the same thing. To add to the confusion, there are a few more fine points here, such as the distinction between http://www.wikidata.org/wiki/Special:EntityData/Q1000 and http://www.wikidata.org/wiki/Special:EntityData/Q1000.nt (the latter being the URI/URL of a specific serialization of the description), or the fact that depending on the Accept header you send, you may be redirected to https://www.wikidata.org/wiki/Q1000 (which is the canonical URL for the HTML representation of the description of Gabon). But in the end, just keep in mind that the thing as such needs a URI distinct from the document's URL. -- daniel