Covering posts from 0800 ET September 13 to 0800 ET September 14. Sources: 166 geospatial feeds.
Quiet day across the feeds — here are the highlights. Twelve posts landed in the window, but a QGIS plugin release, a MapTiler product update, a Trimble teaser video, a personal travel post, an off-topic political newsletter and a thin Mappery photo caption filtered out as noise, leaving six substantive posts. Two of those six point at the same thing: cloud-native geodata publishing keeps getting easier to ship. Chris Holmes walked through the personal history behind Portolan, a serverless SDI for GeoParquet, COG and Zarr, while MappingGIS covered MVT Server, a new Rust-based open-source tool that serves vector tiles straight out of PostGIS for use with MapLibre.
Why this matters: Landscape context tracks cloud-native infrastructure maturing from "interesting format" toward practical tooling, citing CNG's geoparquet-io release. A Tier 1 voice and an independent Spanish-language blog shipping serving tools on the same day and the same theme is more evidence that maturation is broadening past the usual English-language infrastructure crowd.
Note: Volume was very low across the window. Six posts merit selection; nothing here is padding to reach a target count.
1. The road to Portolan — Spatialists Chris Holmes tells the personal story behind Portolan, a serverless spatial data infrastructure meant to make cloud-native formats like GeoParquet, COG and Zarr the default way to share geodata. It's a rare first-person account from one of the people actually building the cloud-native stack, not just writing about it, and it frames AI as part of the bet on making that infrastructure accessible. → Read it
2. MVT Server: publica teselas vectoriales desde PostGIS con MapLibre — MappingGIS MappingGIS covers MVT Server, a new open-source Rust project that serves vector tiles directly out of PostGIS for use with MapLibre, cutting a chunk out of the usual tile-serving stack. It's concrete cloud-native tooling from a Spanish-language voice the mostly English feed ecosystem rarely surfaces. → Read it
3. The Evolution of Commercial Satellite Imagery in the U.S. Government — Project Geospatial Project Geospatial traces seven decades of the US commercial satellite imagery ecosystem, crediting the intelligence officers, researchers, risk-taking founders and procurement officials who built it rather than just the companies involved. It's a rare long-form historical treatment of how government demand actually shaped a commercial industry, sitting squarely in the sovereignty and defense conversation running through the feeds all year. → Read it
4. Closing the loop: GeoAI from scan to decision — Spatial Source Spatial Source reports on CSDILA researchers building an AI process that runs real-world cityscape geospatial assignments end to end, from scan to decision, rather than stopping at a demo. It's a small but concrete data point in a GeoAI conversation that landscape context flags as heavy on announcements and light on deployed systems. → Read it
5. What's in a name? Naming things in OpenStreetMap — OpenStreetMap Blog The OpenStreetMap Blog digs into how the project's name= tag works, noting that volunteers have mapped 115 million names worldwide and walking through how local-language naming actually gets decided on the ground. It's a plain-language explainer on a piece of OSM's data model that shapes how the map reads in every country it covers. → Read it
6. GeoGuessr Gets an Antipodean Time Machine — Maps Mania Maps Mania rounds up two new GeoGuessr spinoffs, including TimePin, which asks players to guess not just where a photo was taken but when, using a time slider alongside the usual map pin. It's a lighter pick, but it's the kind of daily web-mapping curation the blog has built its reputation on. → Read it
Powered by Neptune