GeoJSON
PulseAugur coverage of GeoJSON — every cluster mentioning GeoJSON across labs, papers, and developer communities, ranked by signal.
4 day(s) with sentiment data
-
AI agents lose instructions and code due to context window compaction
AI agents, including those from Anthropic and OpenAI, are losing critical instructions and data due to a process called context window compaction. This occurs when conversations become too long, and the AI summarizes ol…
-
New framework automates urban renewal with GeoJSON edits
Researchers have developed CEAE, a hierarchical agentic framework designed to automate urban renewal tasks by translating natural-language instructions into machine-executable GeoJSON edits. This system breaks down comp…
-
Simon Willison builds GeoJSON to PNG converter tool
Simon Willison has developed a web application that converts GeoJSON shapes into rendered PNG images. This tool was created because Willison found it faster to build a solution than to search for an existing one that me…
-
AI tools assist in creating GeoJSON map viewer for local boundaries
Simon Willison documented his process of creating a GeoJSON map viewer, utilizing AI tools to generate boundary data for local political districts. He collaborated with AI models like GPT 5.6 "Sol" and Claude Code, refi…
-
GPS distance calculations often wrong due to latitude/longitude degree differences
A common error in calculating distances between GPS coordinates stems from treating degrees of longitude as equivalent to degrees of latitude. While a degree of latitude is consistently about 111,320 meters, a degree of…
-
Mapbox enables live maps in AI agent chats with MCP Apps
Mapbox has introduced MCP Apps and a new tool called `render_map_tool` to enhance AI agent interactions by allowing live maps to be displayed directly within chat interfaces. This addresses the limitation of traditional…
-
Choropleth maps enhance foundation model spatial reasoning, study finds
A new study published on arXiv explores the utility of choropleth maps for enhancing the spatial understanding capabilities of foundation models. Researchers developed ChoroplethMap-Bench, a benchmark comprising 2,400 s…
-
New VecLang model treats remote sensing mapping as text generation
Researchers have introduced VecLang, a novel approach that treats remote sensing vector mapping as a structured text generation problem. This method encodes geospatial entities like buildings and roads into a GeoJSON-li…