The dashboard will generate a list of edited items, which can be downloaded. We can then do a more nuanced analysis of them, beyond what is done in the dashboard.
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Resolved | Jopparn | T239592 Design a back-up campaign aimed at the community (FindingGLAMs Challenge) | |||
| Resolved | Jopparn | T243901 [Feb 24–28] Wrap up the FindingGLAMs Challenge | |||
| Resolved | Alicia_Fagerving_WMSE | T243899 [Feb 26] Compile a results list for the Challenge | |||
| Resolved | Alicia_Fagerving_WMSE | T243992 [Feb 26] Analyse edited Wikidata items |
Example query to plug the data into to visualize the items on a map. The map layers are made from manually assigned variables, and the item lists can be made in a spreadsheet using the exported data from the dashboard.
Example query to analyze the types of institutions edited. As an attempt at clarity, it only counts P31's that are a subclass of GLAM – some items are both a glam and a building, for example.
Hm. No idea if it's on purpose or not, but when you export the data, it includes all the edits made by the users within the time frame – both on tracked and untracked items. That's despite the fact that the tracking is accurate on the live dashboard. That's quite unexpected. If that's how it's supposed to work, the analysis will have to include an extra step to cross-check the export with the Petscan results to make sure only relevant items are included (T245013)
Edit: Issue has been fixed :)
I have manually triggered a data update, which should be done within the next couple hours, just to make sure everything is up to date. Will download and compile the results once that is done.
collected info about countries:
https://meta.wikimedia.org/wiki/FindingGLAMs_Challenge/Created_items_countries
https://meta.wikimedia.org/wiki/FindingGLAMs_Challenge/Edited_items_countries
Which was included in our final report.