Wikidata:GLAM/Oxford

From November 2017 to October 2019, User:MartinPoulter worked on a project at the University of Oxford to 1) share data from the university's GLAMs and 2) use these data sets to build proof-of-concept applications to expose these data and provide new pathways into and across collections. This placement follows on from the year Martin spent working on various projects across the University, which in turn followed on from a year as Wikimedian In Residence at the Bodleian Libraries.

Martin has also been advising the Oxford Linked Open Data (OxLOD) project on using Wikidata to reconcile existing data sets.

The Wikidata project has an entry on the GLAM Search & Discovery site with more about the goals of the project.

Final report edit

Wikidata search & discovery project for Oxford GLAMs: final report

The final report is an 18-page PDF with screenshots.

Data sets from this activity edit

Ren Xiong, a Chinese artist whose work is represented in the Ashmolean

Eastern Art of the Ashmolean Museum (Q636400) edit

3,409 items . 30,554 statements (mean 9.0 per item) as of end of May 2019 (Recent Changes)

Collection Explorer is a Wikidata-driven application I've written to browse this data set, along with some items from the Bodleian, and other databases including the Oxford Research Archive and Early Modern Letters Online.

List of an artist's works in Listeria

PAWS notebook with some queries for this data set

Ashmolean entry in Sum of All Paintings

Artworks of the botanical illustrator Ferdinand Bauer (Q79012) edit

The intention is to connect this set of more than a thousand artworks with samples from the University's Herbaria and with records of the Sibthorp-Bauer expedition of 1786-88, connecting the works to places, dates and scientific data.

2,106 works as of the end of April 2019. 12,493 statements (mean 5.9 per item)

Shah Jahan, a Mughal emperor depicted by items in both the Bodleian and the Ashmolean

Manuscripts and paintings in the Bodleian Library (Q82133) edit

This set is growing gradually as the Bodleian puts more complete digitised works online. This set builds on some items that were originally imported from Wikipedia.

2002 items as of the end of March 2018; 14,858 statements (mean 7.4 each)

Art works in the Pitt Rivers Museum (Q1456119) edit

43 items added in June 2018 (drawings and lithographs); 294 statements

astrolabe (Q164992) edit

History of Science Museum (Q6941088) has shared some of its data about its collection of 164 astrolabes. In early September, it shared 164 images on Commons. Astrolabe Explorer is a proof-of-concept application I've created to browse data about astrolabes from multiple collections. 1574 statements (9.6 statements per item) as of the first week of September 2018.

In November, Science Museum (Q674773) shared a data set of their 25 astrolabes which Martin added to Wikidata.

The National Museum of American History (Q148584) already had a data set of 33 astrolabes available on its web site, so that has been included in December 2018.

The Adler Planetarium (Q358052) shared data and images about their collection of 97 astrolabes. These have been added in July 2019.

Other data sets edit

Some data sets were added to Wikidata as part of the placement funded by the university's IT Innovation Fund.

Oxford doctoral theses imported from Oxford Research Archive 3,239 as of the end of 2017
Early Modern Letters Online: 6,812 person identifiers and 3,335 location identifiers in Wikidata as of the end of January 2019. See User:MartinPoulter/queries#Early_Modern_Letters_Online
Electronic Enlightenment person identifiers: 3,317 in Wikidata from an import in 2017. See User:MartinPoulter/queries/ee#Maps
The 18th century poetry archive have shared a data set, but I've not yet had time to add this to Wikidata.
The Beazley Archive Pottery Database (BAPD) have shared a data set of around 100,000 items, not all of which are suitable for Wikidata. This needs a chunk of time to be worked through and reconciled.

Events edit

"Teaching SPARQL as a Foreign Language" 20-minute presentation at WikidataCon 2019, Berlin, 26 October 2019
"Wikidata as a Platform for Linked Open Data About Collections" 15-minute presentation at Museums+Tech 2019, "Openness", 18 October 2019
"Adding to the global web of knowledge: Wikipedia, Wikidata and open platforms" 2-hour Presentation at at Berlin Summer School on Open Research, 18 September 2019
Martin was a panellist in a webinar for the Wikidata Affinity group of the Linked Data for Production (LD4) project, 30 July 2019
"Wikidata: Knowledge Representation the Easy Way": lecture at Digital Humanities Summer School (different from the 2018 lecture despite the same title), 24 July 2019
"Wikidata and GLAMs": 1 hour talk for library and museum staff in Oxford University, 3 July 2019
Keynote presentation and workshop on Wikipedia for education in the digital humanities, Maynooth University, Ireland, 24 May 2019
Half-hour presentation on the Wikidata Query Service at the PastNet hackathon hosted by The Oxford Centre for the Humanities (TORCH), 11 May 2019
Talk on "The Confluence of Open" and introduction to Wikipedia editing for University of Oxford Staff, 4 March 2019 (Write-up by Juliet Ralph for Oxford Open Access blog)
"SPARQL as a Foreign Language", workshop for University of Oxford Staff, 28 February 2019
Short presentation at GLAM Collections Committee meeting, January 2019
"Wikidata at Oxford": presentation in webinar for OCLC Wikimedia Interest Group, 26 November 2018
"Astrolabe Explorer- the power of Wikidata to explore collections": presentation (with Silke Ackermann) at the university's GLAM Digital Showcase, 21 November 2018
"Things not Strings": presentation to IT Services staff about Wikidata and Knowledge Graphs, 1 August 2018
"Wikidata: Knowledge Representation the Easy Way": lecture at Digital Humanities Summer School, 5 July 2018
"The Convergence of Open: Open Access, Education, Data and Culture", Seminar on Reproducibility and Open Research, 1 May 2018
"Wikidata, the Semantic Web and the Emerging Global Knowledge Base" 15-minute presentation at OER conference, April 18
"Wikidata for researchers" 5 minute lightning talk at Workshop on collaborative research in Digital Humanities, Museums and Collections, and Publishing, University of Oxford
"The global knowledge base: open data about everything", TEDx, University of Bath, 24 March 2018
"Wikidata: Knowledge as a Service" Presentation to staff at Oxford e-Research Centre, 28 February 2018

Advocacy edit

Bodleian Digital Library blog: "Welcome Japan Search to the web of Linked Open Data", May 2019
"The hyper-engaged volunteers helping Oxford GLAM" Published on the GLAM Digital Strategy site, May 2019
YouTube video: "Wikidata and Oxford's GLAMs", April 2019
Wikimedia UK blog: "A call to action: Wikidata-fy your Commons photos", March 2019
Bodleian Digital Library blog: "What Wikidata offers Oxford’s GLAM Digital Strategy", March 2019
Bodleian Digital Library blog: "Build your own Digital Bodleian with IIIF and SPARQL", March 2019
Collection Explorer briefing, published on the GLAM Digital Strategy site, March 2019
Bodleian Digital Library blog: "Making Wikidata visible", January 2019, republished by Wikimedia France as "Rendre Wikidata Visible"
Bodleian Digital Library blog: "Detailed depictions with IIIF, Wikidata and Wikimedia Commons", December 2018
Bodleian Digital Library blog: "Wikidata and GLAM catalogues: another round-up", October 2018
Bodleian Digital Library blog: "A global collection of astrolabes in linked open data", August 2018
Bodleian Digital Library blog: "Wikidata and GLAM catalogues: a round-up", July 2018
Bodleian Digital Library blog: "Translating a blog post into structured data", May 2018
Bodleian Digital Library blog: "Some ways Wikidata can improve search and discovery", February 2018
Bodleian Digital Library blog: "A Reconciliation Recipe for Wikidata", February 2018
Bodleian Digital Library blog: "Semantic data and the stories we're not telling", November 2017
Bodleian Digital Library blog: "Reconciling database identifiers with Wikidata", October 2017
Bodleian Digital Library blog: "Resource discovery and Wikidata", June 2017
Oxford University Museums Partnership blog: "Wikidata: the new hub for cultural heritage", January 2017

Similar work elsewhere edit

Other institutions in the cultural sector are also engaging with Wikidata for search and discovery.

"Crotos is a search and display engine for visual artworks based on Wikidata and using Wikimedia Commons files."
The Wikidata Scholar at the National Library of Wales was one of the inspirations for this project. The NLW has since shared data about 5,000 portraits: Wikidata has allowed them to disambiguate artists and to create multiple visualisations of the collection.
The Flemish art museums Wikidata project is an example of combining catalogues from different GLAM institutions.
The Museum of Modern Art in New York (MOMA) are using Wikidata to provide information about the artists in their database, drawing in text extracts from Wikipedia and links from the Getty Union List of Artist Names (ULAN)
The National Galleries of Scotland are similarly connecting their artist records with Wikidata, bringing in extracts from Wikipedia and ULAN.
Dan Scott at Laurentian University library is integrating Wikidata with the library's cataloguing interface. See Stacy Allison-Cassin & Dan Scott's paper in The Code4Lib Journal.
- "libraries and archives should invest their limited resources in enriching Wikidata, a centralized data repository, to maximise the visibility of those entries and the reusability of that data in the world at large... and then pull the data back into our own repositories"
VIAF (the Virtual International Authority File) is importing Wikidata links and using Wikidata as a source for alternative names for people and organisations, especially from non-English sources.
A consortium of US libraries created "Project Passage", a joint linked data platform based on Wikibase, which makes use of some Wikidata identifiers. They report, "The building blocks of Wikibase can be used to create structured data with a precision that exceeds current library standards."
Research commissioned by Europeana recommends that "The best datasets to use for name enrichment are VIAF and Wikidata". It reports Wikidata's strength as being its multilinguality, providing translations of labels. OCLC also report "We are merging descriptions of translated works with corresponding Wikidata descriptions to obtain the language and script of the original work, the source and target languages, and the names of translators."
Karen Smith-Yoshimura at OCLC has written a short but great blog post with quotes from institutions including the Smithsonian and the British Library.
Wikimedia Sweden is being funded by the National Library of Sweden for a project on "Strategic Inclusion of Library Data on Wikidata".
Stanford University Libraries are advertising for a Wikimedian In Residence specifically to work on library data in Wikidata.
Wikimedia Sweden, in partnership with UNESCO and Wikimedia Foundation, is running FindingGLAMs, a project to use Wikimedia platforms to build a shared database of global GLAMs and share data and images from their collections.
The National Library of Portugal is automating the upload of catalogue data from its open data portal to Wikidata.
Somebody tweeted from a conference that the Bibliothèque nationale de France (National Library of France) have installed their own instance of Wikibase for bibliographic data. I'd love to hear more!
In early 2019 the Sloan Foundation approved a $495,000 bid to improve Scholia and thereby improve the discovery of scientific literature.
The 2019 paper "Museum Collections on Wikipedia: Opening Up to Open Data Initiatives" argues that museums should progress beyond short-term Wikimedia collaborations towards workflows that make Wikidata and other Wikimedia sites integral to the normal work of giving access to museum content.
In May 2019, Library of Congress started displaying Wikidata links on over a million of its authority file records
The Always Already Computational: Collections as Data project has some recommendations for libraries relating to Wikidata, as well as some reports on present activity.
The Linked Data for Production: Pathway to Implementation (LD4P2) project has a Wikidata Affinity Group led by Wikimedian In Residence Hilary Thorsen, based at Stanford.