*Governance*

From rphyrin’s summary #OpenStreetMap 2026 Local Chapters and Communities Congress, two points:

“How can we adapt #tagging practices to local realities?” Yes real lack of tags for Global South, and locals rarely suggest them because not used to the process. This needs to get started!
https://www.openstreetmap.org/user/rphyrin/diary/408430

“Documentation mostly in English.” No, I spent nearly three years creating a 6-language #MOOC (EN, ES, FR, IT, PT, ZH) #UNMappers #Learning Hub https://maps.un.org/learning/

*Scientific papers* 1/2

Interesting study by @heigit on #AI-assisted integration of road #data into #OpenStreetMap, which shows the more corrections made using #HumanInLoop, the less #transparency and #traceability regarding #DataQuality
Personally, always refused to switch from being a skilled craftsman to a controller of standardized products, but here’s another argument against introducing #AI #geodata into #OSM. Solution: #mashup #OSM + AI data to fill in the gaps.
https://www.tandfonline.com/doi/full/10.1080/24694452.2025.2589286

*Scientific papers* 2/2

A #lowcost #deeplearning methodology integrating #OpenStreetMap for #urban planning tasks across multiple domains. #OSM #vectors serve as reference masks for #satelliteimagery to enable predictions via #CNN and #GNN. Accuracy >90%.
New #remotesensing products require #vector #geodata. Hence the value of investing in the completeness and consistency of #OSM data
https://www.
nature.com/articles/s41598-026

*Tools*

Duckdb-osmium allows you to read #PBF files directly into #duckdb without prior conversion. Doesn't say whether supports other types #osm files (history, changesets), but this needs to be tested
https://
github.com/jake-low/duckdb-osmium

Project NOMAD aims to provide an offline server containing various knowledge and educational tools. Unlike https://
labriqueinter.net/ (which has different goals), this project currently lacks description of use cases.
https://github.com/Crosstalk-Solutions/project-nomad