I guess then the question is how we will handle the articulation between class and queries then, we have a ''subclass of" property which uses items as value. The same relation exists beetween queries.
In my view the answers of the query as instances and explicitely stating that an item as instance of some class are not totally redundant, at least in reality : we can have a situation in Wikidata where someone just entered the instance of value on an item without setting other properties which would make it a result in the corresponding query.
This would be an inconsistency for a OWL reasoner maybe, but it's a hint that there is missing properties or omething wrong in the database for Wikidatans ... So it's not totally redundant in term on vandalism robustness or error handling : if a change is made the put an item out of a query result but the equivalent instance of statement is set, we can have a hint in a database report. As Wikidata is a wiki, I think this is something that is worth thinking of.
2014-03-06 20:42 GMT+01:00 Denny Vrandečić vrandecic@gmail.com:
I think the issue is simply a problem of terminology right now.
Wikidata as it is does have no support for "Classes".
In the future, there will be Wikidata Queries.
A Wikidata Query is similar to a class description in OWL.
There is no need to create an item "Douglas Adams Novels". You can, though, have a Query "Douglas Adams Novels", and change its label, etc. Wikidata Queries will also have labels, an ID, etc.
Does this make sense?
On Thu Mar 06 2014 at 11:33:36 AM, Thomas Douillard < thomas.douillard@gmail.com> wrote:
Hi Lydia, in languages such as OWL, a class of item can be defined by a predicate on the properties and values of the item. Will we be able to link a class Qitem to a Wikidata complex query in the corresponding namespace ? For example if we define a class "Douglas Adams Novels" whose instances are the members of the example query you gave ?
2014-03-06 19:24 GMT+01:00 Lydia Pintscher lydia.pintscher@wikimedia.de :
On Tue, Mar 4, 2014 at 9:01 PM, David Cuenca dacuetu@gmail.com wrote:
I would like to make you aware of this RFC started by Gerard:
https://www.wikidata.org/wiki/Wikidata:Requests_for_comment/Define_lists_on_...
It is interesting because in the end, what is the difference between a
list,
a category, and a query? Not much, really.
I'm curious to know if the approach taken with queries will be the
same as
the WDQ http://tools.wmflabs.org/reasonator/?&q=6573995
Items like "List of..." or "Category:" would have some use, but the development notes don't state if this is the intended path https://meta.wikimedia.org/wiki/Wikidata/Development/Queries
Any thoughts about it?
I've been trying to understand the RfC 3 times now and still fail. So I can't answer your questions unfortunately.
The short and simplified version of how complex queries will work:
- someone defines a query on a page in a special query namespace (eg
"everything that has author = Douglas Adams")
- the result of the query is a list of items matching the query
- the Wikipedias can include the result of the query and visualize it
in certain ways on a page. (eg the wikitext of the article "List of works by Douglas Adams" would have a call to include the query result from Wikidata)
Cheers Lydia
-- Lydia Pintscher - http://about.me/lydia.pintscher Product Manager for Wikidata
Wikimedia Deutschland e.V. Tempelhofer Ufer 23-24 10963 Berlin www.wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e. V.
Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 Nz. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l