WikiConference North America 2019: Reliability

Attendees of WikiConference North America 2019

Over the weekend I attended WikiConference North America in Cambridge, Massachusetts. This was my fourth time participating in this meeting, which is a wonderful gathering for Wikimedians as well as librarians, educators and others interested in open access to information. This meeting is purposefully expansive, including colleagues from Mexico, the Caribbean and Canada. It was wonderful to see so many Caribbean participants, possibly more so even than from Mexico or Canada (likely due to the emergence of a new and lively Wikimedians of the Caribbean User Group ).

This year the conference was held in conjunction with the Credibility Coalition and featured a “credibility summit” with participants from Google, Facebook and Microsoft alongside members of the Wikimedia Movement. This convergence facilitated necessary and timely discussions on credibility, reliability and the role that these organizations and communities play in combating fake information on the internet.

There was a contingent of librarians / Wikibrarians attending the conference, several talks that touched on Wikicite, and many talks on Wikidata.  From my perspective, notable talks included:

  • Presenters from Vanderbilt University Library walked through how they are considering using Wikdata (or Wikibase) as a Research Information Management (RIM) system. [notes]
  • Former OCLC Wikpedian-in-Residence and OCLC WebJunction instructor Monika Senjul-Jones gave a lightning talk on our recent Wikipedia + Libraries: Health and Medical Information course.
  • Will Kent of WikiEdu facilitated a discussion on Building a Wikidata Curriculum. This was fed by lessons learned, both by Will and his WikiEdu collaborators and others in the audience who have been teaching Wikidata to others. The session etherpad has a number of good resources for teaching Wikidata.
  • A Harvard libraries panel on digital humanities resources / “non traditional scholarship” included discussion about whether / how these resources might be used in a Wikipedia article and how you might use Wikidata to describe / model them. [notes]
  • A presentation from University of New Mexico explored how Wikipedia might stand in for electronic resources (specifically, to let selectors do analysis on children’s literature). Could they use Wikipedia articles instead of subscription resources, particularly when selectors are interested in finding books that center non-dominant cultures? [slides | article]. I’ve been looking at lists like 1000 Black Girl Books and those featured on We Need Diverse Books and have been thinking about how these resources show up (or, mostly don’t) in Wikipedia and Wikidata, so this is definitely a topic I’m interested in.
  • CiteUnseen is a Wikipedia user plugin that shows what sources are used to cite a Wikipedia article (books, websites, newspaper articles, etc.), and also flags sources that are from questionable sources.

I facilitated a discussion on “gap” projects and what tools / techniques those projects use. Some projects create simple crowd-sourced lists. Others leverage out-of-copyright topical encyclopedias or biographical dictionaries and then push structured data gleaned from transcriptions of these sources into Wikidata and then use Listeriabot to generate lists from them (here is one example from the Women in Red project). Other projects like Art+Feminism are focusing effort on articles that already exist but that are at risk, using a combination of Wikidata and information from Wikipedia articles. Overall I have been impressed by how many gap projects leveraging Wikidata to identify and prioritize work.

The event was funded by the Credibility Coalition, Craig Newmark Philanthropies, and the Craig Newmark School of Journalism, along with support from the Knowledge Futures Group, MIT Open Learning, and many greater Boston area arts organizations. Many thanks are due to the funders, the many volunteers and the program committee for making this fun and thought-provoking meeting possible. I look forward to next year’s event, which will be hosted by Wikimedia Canada.