Wikidata

Wikidata:Property proposal/cites

< Wikidata:Property proposal

cites edit

   Done: cites work (P2860) (Talk and documentation)
DescriptionCitation from one document to another, e.g., one scientific paper to another scientific paper
Representscitation (Q1713)
Data typeItem
Domaindocuments: scientific articles, books, legal texts, ...
ExampleScientific citations in Wikipedia (Q21172284) → Internet encyclopaedias go head to head (Q24239902)
Robot and gadget jobsProbably not
Motivation

During WikiCite there has been some discussion of such a property. I do not know what to think about it. It could potentially "blow up" Wikidata with millions of documents each citing tens/hundreds of other documents. Nevertheless I think it could be interesting. By having a "cites" we can build citation networks and list relevant similar articles based on co-citation network. It is already possible to build co-author networks. — Finn Årup Nielsen (fnielsen) (talk) 06:13, 26 May 2016 (UTC)[reply]

Clarification: I support the creation of the property and am not afraid of flooding Wikidata. I was just expressing the fear that others could be afraid of flooding Wikidata. — Finn Årup Nielsen (fnielsen) (talk) 08:43, 26 May 2016 (UTC)https://www.wikidata.org/wiki/Wikidata:Property_proposal/Creative_work#citerer[reply]
Discussion
  •   Support This is a very much needed property for representing basic relations between items that represent sources, per Wikidata:WikiProject_Source_MetaData. I want to strongly challenge the suggestion that the sheer creation of a property will result in flooding Wikidata with millions of items. How to create entries on Wikidata and what's appropriate in terms of data imports has little to do with the usefulness of the property. --Dario (WMF) (talk) 08:40, 26 May 2016 (UTC)[reply]
  •   Support I support the creation of this property. I agree that it is important to be able to represent citation relationships between pieces of scholarly literature --Hweyl (talk) 08:53, 26 May 2016 (UTC)[reply]
  Support I think this property is certainly useful to have. Of note, the most relevant ontology here &mdash the Citation Ontology (CiTO) — has an element "cites" that refers to basically the same thing proposed here, though it is more focused on scholarly publications specifically. --Daniel Mietchen (talk) 08:45, 26 May 2016 (UTC)[reply]
  •   Support I support the creation of this property. It is inconceivable that a property representing relationships between the foundations of knowledge would not exist. Porteclefs (talk) 09:43, 26 May 2016 (UTC)[reply]
  •   Support This is an essential property for mapping the literature network which would be very valuable to have as part of the project.--Konrad Foerstner (talk) 08:51, 26 May 2016 (UTC)[reply]
  •   Support This adds power to modeling bibliographic relations on WIkidata --Andrawaag (talk) 08:53, 26 May 2016 (UTC)[reply]
  •   Oppose we already had proposals for "mentions" and the like. This seems too granular.
    --- Jura 08:58, 26 May 2016 (UTC)[reply]
    For reference
    * Wikidata:Property_proposal/Archive/13#mentions
    * Wikidata:Property_proposal/Archive/33#mentions
    * Wikidata:Property_proposal/Archive/45#mentioned_in
    * Wikidata:Property_proposal/Archive/51#mentions
    ·addshore· talk to me! 09:07, 26 May 2016 (UTC)[reply]
    •   Comment For scientific articles the idea of a "citation" is a fairly confined idea. "Mentions" is a somewhat general ("hazy" to me), e.g., The Ugly Ducking "mentions" a "swane". Would that go into it the property? "cites" would corresponds to http://purl.org/spar/cito/cites as mentioned above. — Finn Årup Nielsen (fnielsen) (talk) 09:39, 26 May 2016 (UTC)[reply]
      •   Comment CiTO, the Citation Typing Ontology is far more granular, and in my opinion what we ultimately need. A mere 'cites' does not distinguish 'disagreesWith', 'supports', or 'citesAsAuthority' (i.e. citation like 'because he wears a nice hat'). CiTO does that. However, the world has not moved there yet, and cito:cites is the standing practice of tracking provenance of scholarly knowledge. Egon Willighagen (talk) 11:33, 26 May 2016 (UTC)[reply]
    •   Comment It is certainly true that a citation is an instance of a mention. It is also true that a citation is quite granular. However, the granularity is, in fact, quite meaningful. When a scholar cites prior knowledge, she is engaging in a very specific and explicit act that signals direct recognition of the influence upon current research of the cited item on the surrounding text. In this way, citations are not mere mentions. I can, in fact, imagine clear distinctions between citations and mentions in the scholarly literature. For instance, though Albert Einstein explicitly denies the veracity of many quantum state phenomena, no practicing physicist would explicitly cite this denial. Rather, in motivating work, she might opt to merely mention that this or that well established and observed phenomena was, at one time, denied by the world's most famous scientist. This mention is certainly a statement of fact. But, what it lacks is the granular force of idea propagation along an explicit citation. Rather, it is, in a word, conversational and, as such, is not preserved in the more explicit networked system of citations. Porteclefs (talk) 09:40, 26 May 2016 (UTC)[reply]
  •   Support for a specific reason. Many of the papers NIOSH (and our affiliates) publish are primary in nature and therefore unlikely to be cited directly on Wikipedia. However, review articles that use these papers are more likely to be cited. Modeling these relationships is a crucial component of determining (a) which papers are cited on Wikipedia and (b) of these papers, how many cite our research. This will help us measure our impact on Wikipedia. James Hare (NIOSH) (talk) 09:21, 26 May 2016 (UTC)[reply]
  •   Done cites work (P2860) --Tobias1984 (talk) 09:30, 26 May 2016 (UTC)[reply]
  •   Comment less than 3 hours from initial proposal to creation of a property is very far from good practice as it does not allow most people the opportunity to read, understand and comment on the proposal. Yes there was strong numerical agreement here, but there was no diversity of input and no ability for anyone to counter any groupthink that may be happening. I don't know whether this proposal has merit or not, but I do know that regardless of merit it should not have been created without several days discussion at absolute minimum. Thryduulf (talk: local | en.wp | en.wikt) 10:09, 26 May 2016 (UTC)[reply]
  • Some four minutes after creation, the property was nominated for deletion. --Yair rand (talk) 10:12, 26 May 2016 (UTC)[reply]
  • Very strong   Support. Formal citations are a key approach for providing provenance. In fact, implicitly Wikidata already uses this daily (in various ways). But as others pointed out, a mechanism is needed to link information sources at a different level: not within a single Wikidata entry, but between entries. Citations is how the world keeps track of the origin of information, and I support the idea that Wikidata follows what the world has been doing for ages already. Egon Willighagen (talk) 11:30, 26 May 2016 (UTC)[reply]
  •   Comment As a result of the issues around this speedy property creation I have started an RFC to try and prevent future controversies. You are invited to comment at Wikidata:Requests for comment/Standards for property proposal discussions, but note that the RfC is about the general case in future, not this specific property. Thryduulf (talk: local | en.wp | en.wikt) 12:14, 27 May 2016 (UTC)[reply]
Retrieved from "https://www.wikidata.org/w/index.php?title=Wikidata:Property_proposal/cites&oldid=344415668"