Commons:OpenRefine/Training 2023-24/Midpoint report

For 2023-24, as part of its support for Wikimedia Commons, the Wikimedia Foundation is funding OpenRefine for bug fixes to its Commons features, for a train-the-trainer program, documentation, and a WikiLearn course. This page contains the midpoint report of this grant, January 2024.

• General info page about this project, including a log of activities
• Dedicated info page about the OpenRefine-Wikimedia train-the-trainer program

Since the start of this grant in July 2023, the following activities have taken place.

Train-the-trainer course

This grant intends to increase capacity in the Wikimedia movement for training in OpenRefine. For this purpose, a train-the-trainer course is organized, with a focus on the Wikimedia Commons features in OpenRefine.

• End August 2023: a general call for participants in this course was published. 84 candidates expressed interest. Due to this high level of interest, 16 participants were selected instead of the originally planned (maximum) 8. The list of participants of this course is available on the course page; criteria that were considered for selection were regional and linguistic diversity, the candidates' existing experience in Wikidata, Wikimedia Commons and OpenRefine, and the candidates' potential to indeed widely continue with training after this course.
• November 2023: The course started, and will run until end April 2024. It consists of roughly two parts:
  • Baseline skill-building (October-December): all trainees must refresh or build their basic knowledge of OpenRefine and Wikimedia Commons by making sure they have done edits and uploads in the past year
  • Practice training and share experiences (January-April): all trainees must give OpenRefine trainings in their own context and share their experiences with the group.

During the course, participants communicate via a dedicated Discord channel which is kept internal to the group, to ensure a safe learning experience. There are also two-weekly group meetings in alternating time zones, with recordings that are shared internally in the group. The course instructor also had 1:1 check ins with each participant to hear about their learning needs and in which areas they need (more) support. These conversations showed that the course curriculum is comprehensive, but several course participants indicated they wanted to learn more about GREL, data modeling, scraping and other advanced tasks, and want to share experiences around the didactics of teaching OpenRefine.

Data collection and feature prioritization

• September 2023: start of data collection. One goal of this grant is to increase and broaden uptake of the Wikimedia Commons features of OpenRefine which were built as part of a 2021-2022 Wikimedia grant. To be able to measure the impact of these features, data is now regularly collected about usage and uptake.
  • The data is available in a Google spreadsheet.
  • A first overview of collected data was published on OpenRefine's forum.
  • At the time of this report (end January 2024), there have been 135 uploaders and more than 200,000 media files contributed to Wikimedia Commons using OpenRefine.
  • One general initial observation: the Wikimedia Commons features in OpenRefine are (initially) mainly used by people working for cultural institutions, or people supporting them, like Wikimedia affiliate staff. Only few files are uploaded by people who can be identified as (only) Wikimedians.
• September-October 2023: the existing users of the Wikimedia Commons features received a request (example) to participate in two surveys: one to learn more about their profile and learning needs, and one in which they prioritized feature improvements (both Wikimedia Commons and general OpenRefine features).
  • Out of the 84 polled users, 21 responded.
  • The results of these surveys were published on OpenRefine's forum.

Improved documentation and a dedicated course on WikiLearn

• January 2024: the documentation pages about OpenRefine for Wikimedia Commons were overhauled, with updated how-tos for batch editing and uploading files with Wikimedia Commons. Based on feedback from the previously surveyed users, the pages include short demo videos with captions that can be translated on Wikimedia Commons.
  Example of a demo video, for preparing an upload schema in OpenRefine:

• January 2024: a beta version of a dedicated online learning course for Wikimedia Commons editing and uploading with OpenRefine was released on the WikiLearn platform. This course will be beta tested by 8 people, and then publicly released. Anyone with a Wikimedia account will be able to follow this course at their own pace.

Looking forward

After this interim report, in the second part of this grant period, the following activities will take place:

• Finishing the train-the-trainer course and listing and certifying the new cohort of Wikimedia-OpenRefine trainers
• The WikiLearn course will be officially published and then translated in three languages (French, Spanish and Portuguese)
• Conversations about continued collaboration and sustainability between the OpenRefine and Wikimedia communities
• Bug fixes to Wikimedia Commons features in OpenRefine