1 of 32

A splash of

Wikidata

2 of 32

Chris Koerner

Community Relations Specialist

Wikimedia Foundation

3 of 32

4 of 32

5 of 32

Not just Wikipedia

Wiktionary Wikiquote Wikibooks Wikisource Wikinews Wikiversity	Wikispecies Wikivoyage Wikidata Wikimedia Commons MediaWiki

This includes projects like Wikipedia, but not just Wikipedia - the most popular project in the movement.

We support all these projects under the mission - “[To] empower and engage people around the world to collect and develop educational content under a free license or in the public domain, and to disseminate it effectively and globally.”

All free, all collaborative. From a free encyclopedia to a free dictionary, to travel guide, and media repository

And others services and projects like our cloud services for open-source projects

To be clear; free as free to access, free to reuse - even commercially, free to contribute, etc. Not free and we’ll collect data on your habits, not free and we’ll shovel intrusive ads at you, not free but you are restricted in how you use it. Truly, genuinely, free.

6 of 32

How we support

Advancement

Product

Technology

Community Engagement

Legal

7 of 32

Obligatory cat gif

CC by SA 4.0, Jeff Elder

8 of 32

You are lucky learning about it now. You are a Wikidata hipster

9 of 32

A free, collaborative, multilingual, secondary database, collecting structured data

10 of 32

History

Little history

Who needs this? Well, at first it was just us.

Wikimedia projects like Wikipedia are available in close to 300 languages. Data - from everything between a birth date and the GDP of a country - was duplicated, wrong, and missing across articles on the same subject. Starting in 2013 the movement set out to centralize data in a single repository, connect it to projects and their articles, and keep the "wiki way" of contributing. It was called Wikidata.

Wikimedians wanting to keep links to other versions of articles about the same topic in different languages.
Infobox data (like birthdates, census data, GDP, etc.
List articles based on data in Wikidata

data goes out of date or is updated - census data as an example. Volunteers would have to manually run around and update that in multiple locations. Who knows where the value appears on-wiki! So this was time-consuming and difficult.

All data is licensed under CC0 or public domain.

“All structured data from the main, property and lexeme namespaces is available under the Creative Commons CC0 License;”

11 of 32

Collaborative

12 of 32

Multilingual

13 of 32

Secondary

14 of 32

Structured Data

SELECT ?item ?itemLabel ?geo ?type ?typeLabel (SAMPLE(?_image) AS ?image) WHERE {

?type wdt:P279* wd:Q3839081 .

?item wdt:P31 ?type ;

wdt:P625 ?geo .

SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". }

OPTIONAL { ?item wdt:P18 ?_image }

}

GROUP BY ?item ?itemLabel ?geo ?type ?typeLabel

15 of 32

How it works

16 of 32

17 of 32

wikidata.org/wiki/Q3100174

18 of 32

19 of 32

How you can use it

Great Chris I get it. Your 5 minute demonstration has convinced me. Now how do I use it? :)

how you can use it - https://www.wikidata.org/wiki/Wikidata:Data_access

Linked Data interface - https://www.mediawiki.org/wiki/Wikibase/EntityData

https://www.wikidata.org/wiki/Q42

JSON - https://www.wikidata.org/wiki/Special:EntityData/Q42.json

P31 “instance of” property Q5 value (human) - seems mostly harmless

api - https://www.mediawiki.org/wiki/Wikibase/API

sparql - best for getting your feet wet (DEMO TIME)- https://query.wikidata.org

birthday today

Baseball players born in St. Louis - http://tinyurl.com/y7skhzf2

For other Wiki data: https://quarry.wmflabs.org/

Also has an API: https://www.mediawiki.org/wiki/Wikidata_Query_Service

R wrapper for WDQS API; send a SPARQL query, get back a data.frame - https://cran.r-project.org/package=WikidataQueryServiceR

Dumps - https://www.wikidata.org/wiki/Wikidata:Database_download - useful for offline use (JSON, RDF, XML)

Bots - https://www.wikidata.org/wiki/Wikidata:Bots

Donate data - https://www.wikidata.org/wiki/Wikidata:Data_donation

20 of 32

query.wikidata.org

21 of 32

How it’s being used

How it's being used

Wikipedia itself!- https://www.wikidata.org/wiki/Wikidata:How_to_use_data_on_Wikimedia_projects#Examples_of_use_cases

https://fr.wikipedia.org/wiki/Modèle:Infobox_Fromage

Placehodler aritcles - beta softare - generates placeholder articles where one doesn’t exist - https://www.mediawiki.org/wiki/Extension:ArticlePlaceholder

Example on the Gujarati Wikipedia (a language native to the western state of Gujarati - 56 million speakers in India) - an entry on Theodore Roosevelt https://gu.wikipedia.org/wiki/વિશેષ:AboutTopic/Q33866

Google - knowledge graph - https://www.google.com/search?q=Urdu+language - information is sources from Wikidata among others

Apple - Siri knowledge panel - (CLICK) Here’s a query asking “What is Nuclear fission?” The “Award recieved” and “Time of discovery or invention” is sourced from wikidata - https://www.wikidata.org/wiki/Q11429

Amazon - Alexa

https://blog.quora.com/Announcing-Wikidata-References-on-Topics

https://www.wikidata.org/wiki/Wikidata:Wikidata_for_authority_control

https://twitter.com/pigsonthewing/status/867270004017909760

22 of 32

What’s next?

23 of 32

24 of 32

Structured Data on Commons

Structured data on commons

Commons is our media repository. Millions of files - images, animations, video, sound. Most unstructured.

All information about a file is wikitext - an open field of whatever. Ok, there’s some structure with templates and categories.

https://commons.wikimedia.org/wiki/Commons:Structured_data

SDoC converts the free media files on Wikimedia Commons to a structured and machine-readable format, so that they become easier to view, search, edit, organize and re-use. To achieve that, the Commons backend is migrated to Wikibase, the same technology as used for Wikida

Idea is that you could define the metadata about a media (copyright, lens used, date taken, location, etc) and the contents (what does the media depict)

Great presentation to learn more:

https://upload.wikimedia.org/wikipedia/commons/6/60/Structured_Data_on_Commons_-_EuropeanaTech_2018.pdf

25 of 32

26 of 32

Lexeme Data

27 of 32

28 of 32

Your Contributions

29 of 32

Wikidata hipster

30 of 32

Thank you!

31 of 32

Questions?

32 of 32

Credits