Showed my Advanced GIS students Claude Code this week. We built this map of code compliance violations in Fort Worth; heatmap giving way to points on zoom. Used the ArcGIS SDK for JavaScript. Took 10 minutes. I believe in Jevons Paradox. But we need to be teaching skills for 2026, not for 2016.

West Texas energy infrastructure identified from satellite imagery using zero-shot image segmentation. Shown: Well pads from simple text prompts; Draw a box around a wind turbine, find the others in the view; Click a solar farm, segment the area.

Half of the United States' population lives in these 144 counties. Explore: walkerke-county-popu...

In Feb 2024, I asked ChatGPT: "What do income levels by neighborhood look like in Dallas, Texas? Make me a map of the results." The result: the first nonsensical image. Today I asked Claude that question - it made an accurate MapLibre map. Wonder where we'll be in 2028...

Quality-of-life improvements to the #rstats {geosam} package: - Interactive swatches let you customize extraction color in the explorer; - The explorer sidebar is now collapsible; - The package installs the stable version of HuggingFace Transformers walker-data.com/geosam

Distributed energy is harder to map than centralized infrastructure, but cutting-edge AI tools can help. Shown: identifying rooftop solar panels in a suburban neighborhood with Mapbox imagery, Meta's SAM3. Useful for utilities, grid planners, and solar prospecting.

Mapped every pipeline in Texas: 368,000 miles of infrastructure at your fingertips. For energy operators, midstream companies, and data center site selectors, understanding pipeline proximity matters. Open data + modern GIS makes it accessible.

It's 2024 5-year American Community Survey release day! The ACS is the best resource for small-area analysis of US demographics. In R, use tidycensus and mapgl to dive into the data right now. Copy-paste the code below for an interactive income map with a legend filter:

{calcite} looking good 💅 🗺️ Esri's Calcite Design System in {shiny}? Yes, please!!! Full-stack spatial in R keeps getting better. github.com/r-arcgis/cal... #rstats

The big story coming out of yesterday's 2025 Census Population Estimates release: international migration declined in EVERY state in the past year. The plot: international migration rates since 2021 (each year covers the previous 12 months). Biggest drops: NY, NJ, CO, TX, UT

The 2025 @uscensusbureau Population Estimates were released today. Some takeaways: - Rapid growth in the Carolinas, Idaho, Texas; - 17 states + PR had more deaths than births; - Florida is #1 for international migration; - SC ranks #1 for domestic migration rate; NY ranks last

I merged a PR for mirai today (fixing an esoteric bug), that came with a performance boost that I'd never have thought existed. That means that on my laptop, the default (with dispatcher) round-trip performance now dips into sub-100 microseconds territory!! Get it now: `pak::pak("r-lib/mirai")`

Nearly 18 million California jobs in 2023, visualized by 2-digit NAICS category with block-level precision. Open data and modern geospatial technology make powerful labor market analytics possible.

Identifying possible empty lots in Fort Worth's South Main district with Meta's Segment Anything Model 3. Draw an "exemplar" prompt, find similar areas in the satellite image. Useful for urban redevelopment, real estate, and more. Try out {geosam}: walker-data.com/geosam

Social media: "if you're not using {Ralph Wiggum|Clawdbot|next week's new framework} you're ngmi" Now you're endlessly trying to optimize your workflows instead of doing the actual work you're avoiding. Agents will get better and better but they don't substitute for agency.

spopt for R implements the Max-P-compact regions algorithm. Draw contiguous regions that optimize both for attribute similarity and compactness. Shown here: regular Max-P with queen weights (left), Max-P-compact with rook weights (right) Learn more: walker-data.com/spopt-r/arti...

LLMs both enable and impede innovation in open source software. They help developers bring previously impossible ideas to life. But they also recommend older solutions to users given their training data. So even if a new innovation is better, users might not know about it.

Nearly 1500 swimming pools found across a 10 square kilometer suburban area, all from a single text prompt. Meta's SAM3 and Mapbox imagery combine for incredible zero-shot geospatial intelligence.

On the blog: interactive legend filters and screenshot export in mapgl. Try out the interactive examples in the post!

Vignette: solving facility location problems with realistic travel-times. Use the r5 routing engine to generate a travel-time cost matrix from OSM data, then optimize facility locations with a lightning-fast Rust backend in spopt-r. Learn how: walker-data.com/spop...

The Huff model is the classic algorithm in retail spatial analysis - and you can now use it in R. Predict: - Which store a customer is likely to visit - Sales potential per location - How new stores reshape the competitive landscape Learn more: walker-data.com/spop...

The Dunning-Kruger effect, Claude Code edition (speaking as a massive Claude Code fan)

A great use-case for Nano Banana Pro: creating #rstats hex logos. It used to take me hours to put these logos together (I'm not a graphic designer), but logos really help with branding R packages. With Nano Banana I feed Gemini the package spec, iterate, and get great results.

It didn't make it into the CRAN release, but I've just implemented draggable legends in the dev version of {mapgl}. It's quite useful when the legend is getting in the way of your analysis! #rstats

New post: learn how to use the new interactive legends and screenshot control in R's mapgl package. These features are great for interactive demos and visualizing custom scenarios without Shiny. The new release (v0.4.4) is on CRAN now, so go get it! walker-data.com/posts/mapgl-...

In the latest mapgl release: a native screenshot control. Workflow: spin up a quick map; use the new interactive legend to filter your data; drag a marker; take a screenshot; share with colleagues. Uses html2canvas internally to capture the exact state of your map view. #rstats

Here's how the new interactive categorical legends in {mapgl} work. Click a category in the legend to turn the corresponding features on/off on the map. Not groundbreaking, but you get this out of the box with one argument. It used to take me forever to code this manually!

Since I started developing the mapgl R package I've envisioned built-in interactive legends. They're finally here in the upcoming package release: a filter slider for continuous legends, and click-to-filter for categorical legends. Blog post coming this week!

My weekly newsletter is out! This week: 🔹 Open Source of the Week - The spopt project by Kyle Walker 🔹 New learning resources 🔹 Book of the week - Building Natural Language and LLM Pipelines by Laura Funderburk ramikrispin.substack.com/p/building-n...

I'm as excited about Claude Code / Opus 4.5 as anyone. It's helped me create things this year that I couldn't have done otherwise. But I notice often it lacks "common sense." Meaning, it makes decisions that "work" but shouldn't be made.

All 13.5 million jobs in Texas, visualized by sector with block-level precision. The just-released 2023 LODES data is one of the most underrated public datasets.

another sweet port from Kyle! A touch of background on the original, for those interested: knaaptime.com/urban_analys...

Announcing spopt-r: spatial optimization for regionalization, facility location, and market analysis in R. My all-time favorite Python package is PySAL's spopt - and until now, it did not have an R equivalent. Get started here: walker-data.com/spopt-r

"How can you be sure that the code the LLM wrote is accurate without manually reviewing all of it?" Here's my advice: 1. Make sure that you can review *outputs*. If something looks wrong, it probably is. 2. Have another LLM review it. I write with Opus, review with Gemini.

Mapping US wind energy infrastructure with hexagonal binning and "fade-on-zoom" effects. Learn how: gist.github.com/walkerke/a4b...

The interactive explorer in R's {geosam} package gets you started with @Meta's Segment Anything Model 3 for object detection from satellite imagery. Shown: sand trap detection on a golf course from text, exemplar, and point prompts. Try out {geosam}: walker-data.com/geos...

The world's time zones, visualized on an interactive, spinnable globe. Try it out- it'll default to your current time and (approximate) time zone: gist.github.com/walkerke/c4a...

#Rstats slides back into the top 10 programming languages. R is far from dead. www.tiobe.com/tiobe-index/

Visualizing US electric power transmission lines. Learn how: gist.github.com/walkerke/a21...

Spatial optimization has applications in nearly every field but it's underutilized. In 2026, I'm working on open source and lightning-fast tools to make spatial optimization accessible. Shown: interactively solving a "location-allocation" problem in an app with instant results.

2025 was a massive year for me as an R developer. This past year I've branched out and focused on bringing interfaces to some of the most exciting geospatial technologies to the R community. Here's a summary:

Working with big spatial data sets in #rstats? You should try {duckspatial}. The dev version of #duckspatial (soon on CRAN) uses #duckdb to perform super fast and memory efficient spatial operations cidree.github.io/duckspatial/... In a benchmark against, {sf}....

Spatial data science in 2026: - Pull demographic data for all ~7k Census tracts in Texas; - Create a territory design solution with spatial and demographic constraints; - Explore the result on a WebGL map; All in seconds. Technology is no longer the bottleneck.

I’ve always been better at ideas than at coding. I’ve had no shortage of ideas, but coding was always the bottleneck. Which is why I’m unbelievably excited about the moment we are in. However, I get why people whose value proposition is coding itself are skeptical and anxious.

This type of post is hardly unique, but I had one of those "Opus 4.5 moments" yesterday. There's a spatial algorithm I've long wanted in R. It's complex, so I wanted a Rust backend for speed. I don't know Rust, and I don't have time to learn it.

My favorite #Python package to use is spopt, a library for spatial optimization. It helps you with: 📊 Facility location planning; 📊 Sales territory design; 📊 Maximizing market share; And much more! Check it out here: pysal.org/spopt/

It’s been a week since I released the geosam package bringing Meta’s SAM3 image segmentation model to R. This week, users have: - Segmented trees in Belgium; - Found paths in rural areas; - Identified boats in harbors; All in R! Try out geosam: walker-data.com/geosam

While some have made the claim that "R is dead!" this year on social media... Claude Code is going mainstream, and guess what language it's using. #LongLiveR #rstats

If you work in geospatial, PMTiles is one of the technologies you need to know going into 2026. This past year, I wrote an R package to help you get started. Learn how here:

Interactive energy infrastructure analytics powered by the {geosam} R package. Find solar farms on a satellite image in West Texas, click points to segment the imagery & extract shapes. Save the result then get back an sf object in a single step with area calculated for you.