Wikidata:WikiProject LD4 Wikidata Affinity Group/Affinity Group Calls/Meeting Notes/2022-05-03

Call details

• Date: 2022-05-03
• Topic: Adding Wikidata QIDs to JSTOR Images, ‘Archives At’: An Opportunity to Leverage MARC to Create Linked Open Data (Daniela Rovida and Jennifer Brcka)
• Presenters: Sharon Garewal (JSTOR), Daniela Rovida (Sharon Garewal) and Jennifer Brcka (University of Notre Dame)
• Link to original agenda with link to recording: https://docs.google.com/document/d/1ji6eTubixBWrAPv7UUV0gxxW7y_lzyZTf4vvzo5Iwiw/edit

Presentation material

• Slides: LD4_Wikidata QIDs to Open Collections.pdf
• Slides:  'archives at'_5_3_22.pdf

Notes

Sharon Garewal, “Adding Wikidata QIDs to JSTOR Images”

JSTOR Open Community Collections Portal - added special collections/primary sources discovery & preservation service to their platform to drive discovery of digital collections for GLAMS.

New to Wikidata, did a workshop with Will Kent to get started.  Began considering a Wikidata project to do with an Open Community Collection, with a hypothesis of whether they could use Wikidata to drive discovery for OCCs.

Framed the project in terms of OKRs:

Objective:  Expand the visibility and usage of Open Community Collections

Key Results:  increase entry points into JSTOR from 0 to 500 by adding JSTOR ArtworkID for an Open Community Collection

Steps:

1. Apply for a JSTOR ArtworkID property (with help from Will)
2. Choose an Open Community Collection to work with
   1. Selected the Metropolitan Museum of Art because some work had already been done with their metadata to link to creator IDs
3. Use OpenRefine to match records
   1. Matching based on MET Object ID already in Wikidata
4. Add hyperlink to Wikidata QID to JSTOR metadata record and republish

Tracked usage for one month (Feb 16-March 16).  

• From JSTOR, 18 users clicked the Wikidata QID link
• 81 users followed links from Wikidata to JSTOR (183 sessions, 20 from mobile)
• International user audience
• 1 download

Next steps:  continue to add more collections, have submitted an RFP to a vendor to expedite work.

Daniela Rovida & Jennifer Brcka, “Archives At’: An Opportunity to Leverage MARC to Create Linked Open Data”

Started contributing to Wikidata during the PCC Wikidata Pilot, and decided to work on an exploratory project where they would add statements to Wikidata linking Notre Dame finding aids to Wikidata items.

Wikidata identified as a potential way to improve the discovery of archival collections, where aggregation of information about collections and holdings are dispersed across multiple databases.  Could creating linked open data in Wikidata for archival collections overcome some of the these problems?

Demonstrated search results in ArchivesGrid for their example beat poet, Ed Dorn.  Searching the title field produced an accurate result, but “class of persons” search returned less reliable results.  Goal to use linked open data to produce uniform relationships among entities and achieve more consistent search results.

Process Overview:

1. Pulling MARCXML from EAD in ArchivesSpace
2. Used MarcEdit to export tab-delimited files with designated MARC fields (1XX, 300, 506, 520, 856)
3. Used OpenRefine to reconcile 1XX fields, pulling in QID where present.  
4. Used a Quickstatements model to add ‘archives at’ statement pointing to ND collections with determined qualifiers (collection size, url)

Provided links to WDQS to show repository holdings, geographic dispersal of archival collections by a creator.  Seeing where other institutions had already added their ‘archives at’ statements, they experimented with adding collection size information as available to get a sense of how a creator’s archival materials were distributed across multiple repositories.

Questions

No questions were listed in the meeting notes.