BHL/Our outcomes/WiR/Status updates/2025-02-21

07 February 2024 - 21 February 2025

Hi, everyone,

Another bi-weekly update on what has been going on in the Wikimedian-in-Residence (WiR) work:

General updates

Using a bot script, many new BHL images were added to Wikidata. The links were based on categorization of volunteers on Wikimedia Commons. Before the bot script, 15869 BHL Images were used on Wikidata, now 21250 BHL Images are used (~5400 more, a 34% increase). These images end up flowing to Wikipedia versions. For example, in the Portuguese-speaking Wikipedia, there were 959 BHL Images before the script, now there are 1449 (~500 more, a 51% increase)
The Structured Data uploads are progressing well. There are many metadata challenges, but the main technical challenges are solved. So far, structured data was added to 2449 files as part of the WiR work and the pace is increasing.

Technical updates

The Commons Impact Metrics dashboard now embedds the results from GLAMorous, displaying how many times files from a category are used in each of the Wikimedia projects. It is also tracking the usage of BHL images across different Wikimedia projects.
Wikimedia Israel has included BHL in their dashboard for statistics, GlamWikiDashboard. The dashboard is known to have a few silent bugs in counts, so results should be taken with a grain of salt, but it is generally useful.
Charmosyna amabilis, charming, lovely and now reconciled back to BHL from the Flickr ID in the metadata
Instead of looking at collections, I have started focusing on individual illustrator, as this accelerates the process of curation. These weeks I went, for example, through the beautiful birds of John Gerrard Keulemann and Jacques Barraband.
I have changed the reconciliation script to pull update OCR information from BHL, including parsing illustrations with multiple names. and to to pull Flickr IDs even when they are not present on Commons using the 2023 BHL Flickr Harvest dataset.
Adding data IA images on Commons without BHL metadata is often quite hard, but with some care it is possible to infer the mappings. For example, this Commons Image is not currently tracked in the BHL category but came from BHL. In the metadata, I could find the IA Flickr ID and the the IA page (155). It was possible, manually, to see that it corresponded to this BHL Page (156) and, consequently this BHL Flickr ID. The offset of page counts (155 to 156, -1) was regular for the category, and so possible to infer the mappings for 40 different images. This approach might be scaled for other lost BHL images on Commons.
I have built a little tool that gets species in a given GBIF range and generates the quickstatements to add taxon ranges to Wikidata. This will be used in a pipeline to infer which images belong to the South America or the Africa categories on Wikidata.

Community updates

During these two weeks, I attended a few engagement/interaction events, including:

On the 14th, I presented some of the work at the Wikimedia Brasil open meeting (available on YouTube, in Portuguese)
On the 18th, I had a chat with Florencia Gratarolla about herbarium specimens and metadata about that on Wikimedia Commons (available on YouTube, in Spanish)
On the 20th, there was a nice Ask Me Anything session in the monthly BHL Staff meeting. Thanks JJ and Colleen for organizing it and everyone that attended it!
Together with Giovanna Fontenelle, from the Wikimedia Foundation, we are planning 3 engagement events with the Portuguese-, Spanish- and French- speaking Wikimedia communities, including live translation to English. These are likely to happen in the last week of March.

And that is about it! If you have any questions or comments, just let me know!