This is the first monthly digest. The rule: two paragraphs, no slide decks, no meetings requested. What shipped, what didn't, what's next.
What shipped
The L7014 pipeline is complete and archived. 500+ US Army maps of Vietnam are georeferenced, datum-corrected (Indian 1960 → WGS84), and served over IIIF. The propagation algorithm — using seed maps to extrapolate corners across the regular grid — eliminated manual GCP placement for roughly 90% of the series. This work is done; the pipeline code has been retired now that the series is fully processed.
We were featured in Saigoneer (January 2026), Vietnam's leading English-language culture publication. The coverage framed the project clearly: we're not building a map app, we're building the spatial memory of a city most of the world has only seen in wartime. The response from the Vietnamese diaspora community was immediate and warm.
What didn't ship
The knowledge graph is fully designed (schema, predicates, temporal encoding, source types) but not yet built. Community tracing UI exists as a label studio but lacks the gamification layer needed to attract contributors at scale. The 3D pipeline is planned in detail — we've adapted the Morlighem (TU Delft 2021) method to our IIIF output — but Phase 3 starts after Phase 1 footprints are in hand.
What's next
Phase 1 focus: ingest 30–50 colonial maps from BnF Gallica/EFEO, vectorize map features into geometry, build the community tracing UI with Photo Hunter + Cartographer tiers, launch the first photogrammetry missions for 5 landmark buildings (Notre Dame, City Hall, Central Post Office, Opera House, Ben Thanh Market). First public footprint dataset target: 1900 Saigon core.
We're actively looking for: a part-time French-reading researcher to work with ANOM/BnF archival sources, OSM mappers interested in historical Saigon tracing, and Blender artists for landmark 3D modelling. If that's you — or if you know a funder with a confirmed fit (French Institute, EFEO, Wikimedia, Asia Foundation) — say hello.