Max et al,

I'm glad your brought up this opportunity for wikidata to work with OCLC. An issue that I haven't seen adequately discussed on the list yet is how citation/source data is expected to be represented in wikidata. 

There is a great opportunity for wikidata to help fix one of the biggest structured data problems that have afflicted Wikipedia, i.e. the lack of dedicated support for citations, which are currently represented via templates (in the best scenario) or raw links (in the most common scenario). Having citation support built into wikidata with the ability to represent sources themselves as entities, and associate them with the appropriate unique identifiers when available (ISBN, DOI, PMID, ArXiV IDs etc) would allow us to fix many of the outstanding limitations of the current lack of support for citations in Wikipedia. 

A central, canonical repository for all sources used as citations across WIkimedia projects (including WikiData itself) would allow us to:

•  simplify the maintenance of citations in WIkimedia projects
• avoid the proliferation of templates in Wikipedia articles and allow citations to be referred to by unique keys
• expose citation data in both human and machine-readable format (enabling the possibility to export records in various formats like RIS, bibtex, EndNote or format them using CSL [1])
• support collaboration around sources – this is one of the most exciting (and promising) recent trends in collaboration in Wikipedia 
• measure usage of sources across languages and projects (in particular measuring the use of open access/libre materials)
• allow editors to annotate the relation between sources or between sources and factual statements using citation typing ontologies such as CiTO [2]

Some of these goals are the same the WikiCite project is trying to achieve [3] but there are different communities, projects and organizations out there that would benefit for a source repository hosted in Wikidata: just to name a few, acawiki, the altmetrics project, OKFN's open blbliography project.

Citation data could be seeded from data parsed from Wikipedia itself as well as open bibliographic databases such as WorldCat (for monographs) or the CC-BY licensed Mendeley catalogue (for scholarly papers) (licensing issues are important but IMO not insurmountable).

Hope to see a good discussion on this topic at the Wikidata summit/hackathon in the next couple of days.

Dario

[1] http://citationstyles.org/
[2] http://imageweb.zoo.ox.ac.uk/pub/2008/publications/Shotton_ISMB_BioOntology_CiTO_final_postprint.pdf
[3] http://meta.wikimedia.org/wiki/Wikicite


On May 29, 2012, at 7:40 PM, Denny Vrandečić wrote:

Hi Max,

thank you for the pointer! I am very excited about the idea of working together with a project like OCLC! We had a discussion on IRC in the office hour, and I just want to point to the answers there, as soon as the logs are up.

Let us make an appointment for Wikimania, and if you have follow ups, let us know!

Cheers,
Denny 



2012/5/25 Klein,Max <kleinm@oclc.org>

Hello Wikidata Wizards,

 

Phoebe Ayers from the Board recommended I talk to you. My name is Max Klein and I am the Wikipedian in Residence for OCLC. OCLC owns Worldcat.org the world’s largest holder of Library data at 264 million bibliographic records about books, journals and other library items. We would really like to partner with you as Wikidata is being built, in incorporating our data into your project.

 

What we can offer:

·         WorldCat.org metadata http://www.worldcat.org/ .

o   Typically, for any work we have most of the following: title, authors, publisher, formats, summaries, editions, subjects, languages, intended audience, all associated ISBNs, length, and abstract.

·         APIs to this data http://oclc.org/developer/

o   And some other cool APIs like xISBN which returns all the ISBNs of all the editions of book on the input of any single one.

·         Library finding tools

o   When viewing a record on our site, we show you the closest library which has that work, and links to reserve it for pick-up.

·         The Virtual International Authority File (VIAF) http://viaf.org/, which is an Authoritative Disambiguation file

o   That means that we have certified data on disambiguation of Authors

·         WorldCat Identities, an Analytics site http://www.worldcat.org/identities/

o   It gives you for Author metadata and analytics:  Alternative names, significant dates, publication timelines, genres, roles, related authors, and tag clouds of associated subjects.

 

What’s in it for us:

·         We are a not-for-profit member cooperative. Our mission is “Connecting people to knowledge through library cooperation.”

·         Since I work at the research group, for now this is just a research project.

o   If at some point this goes live - and you want to - we’d like to integrate the “find it at a library near me” feature, that means click-throughs for us.

 

The ideas:

There are a lot of possibilities, and I’d like to hear your input. These are the first few that I’ve can come up with.

·         Making infoboxes for each book or author that contains all their metadata.

o   Ready to incorporate into all language projects.

·         Using authority files to disambiguate or link works to their creators.

o   Solving DABs

·         Using our analytics (e.g. author timelines) as Wikidata data types to transclude.

o   Curating articles with easy to include dynamic analytics

·         Populating  or creating works/author pages with their algorithmically-derived history and details.

o   Extremely experimental semantic work.

 

I’m roaring and ready to get this collaboration going. I know Wikidata is at an early stage, and we are willing to accommodate you.

Send me any  feedback or ideas,

 

Max Klein

Wikipedia in Residence

kleinm@oclc.org

+17074787023

 


_______________________________________________
Wikidata-l mailing list
Wikidata-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikidata-l




--
Project director Wikidata
Wikimedia Deutschland e.V. | Obentrautstr. 2 | 10963 Berlin
Tel. +49-30-219 158 26-0 | http://wikimedia.de

Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. 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