If you introduce your students to the idea of Rubber Duck Debugging, then you have to give each of them a rubber duck; I don’t make the rules.

I am super hyped to finally share the first release of plumber2 with all of you. This has been the center of my attention for a big part of 2025 and I hope you'll find it a worthy update to the venerable plumber package. The blog post will tell you more #rstats

Stunning oblique satellite 🛰️ image of #Paris from BlackSky. www.linkedin.com/feed/update/...

If you’re a climate/energy person, I will take your portrait pro bono.

A nice intro to #marimo, the #jupyter alternative codecut.ai/marimo-a-mod... #python #datascience

“Listers, which is streaming for free on YouTube, is like Fear & Loathing for birding" is a great description. This is one of the most fun documentaries I've seen in ages. It's simultaneously crass & elegant, in addition to being an unseriously serious celebration of birds. High rec. 4 bird nerds.

A post by @forrest.nyc (he's always finding cool geostuff!) in LinkedIn brought this to my attention: Temporal Embeddings of Surface Spectra for Earth Representation and Analysis (TESSERA) github.com/ucam-eo/tess... There's also a library to access the data: github.com/ucam-eo/geot... #geospatial

Had a privilege and pleasure to teach terra pkg (created by @roberthijmans.bsky.social et al.) and spatial sampling with R together with Alexandre Wadoux at the #EOSummerSchool2025 at WICC. You can follow the terra tutorial from here: differ.blog/p/processing... 📹: youtu.be/q9sQkg7BeAc?...

Fantastic in-depth article about 19 year old Dominique Malonga - tracking her progress and thought process during her first year in the WNBA.

Equivalence between pairwise correlation and VIF in multicollinearity filtering. Experiment: - Subset df (30k rows, 249 cols) to random dimensions. - Filter using a random max correlation. - Find VIF producing the most similar result to the step above. - Repeat 10k times. #rstats 📦 {collinear}

What a beautiful way to deliver rich spatial data. Many companies could learn a thing or two from this: explorer.audubon.org/explore/spec...

I'm not impressed.

First post on Bluesky! 🌿 Excited to share our new study in Ecography, led by Maxime Lenormand: combining naturalist inventories and satellite data to map plant biodiversity. Here’s a bioregionalization of the flora of France! 🔗 nsojournals.onlinelibrary.wiley.com/doi/full/10....

Our paper now in @nature.com Scientific Data "Global 30-m annual median vegetation height maps (2000–2022) based on ICESat-2 data and ML" doi.org/10.1038/s415... 🌍 stac.openlandmap.org/gpw_gsvh-30m... 📱 global-pasture-watch.projects.earthengine.app/view/gsvh-30m 📰 landcarbonlab.org/insights/glo...

My 2.3% Neanderthal DNA is feeling insulted.

After a bit of fiddling, I finally have a functional Jenkins job to back up ~1TB of Dropbox data in my old but trusty NAS whenever I start my computer. I've used it at work before, but now that I am using it for my own stuff, I can say this out loud: Jenkins is pretty cool!

I still have snapshots of some of the Kepler workflows I worked with during these years. These franken-workflows combined Bash, Grass GIS, R, and even Octave. And ran simulations for months on a few of my lab's computers!

Aha, I found the whole figure!

I found this old 3D representation of a dispersal simulation between populations that I coded during my PhD. It combined species distribution models, cellular automata, and least-cost paths. If my memory doesn't fail me, I used OpenModeller for the SDM, and Grass GIS for the simulation. Fun times!

THE VALKYRIES JUST MADE WNBA HISTORY. 🙌 With their 84-80 win over the Dallas Wings, the Golden State Valkyries are the first expansion team to reach the playoffs in their first season of existence!

Alien Earth's multi-eyed fucktopus is kind of a dick.

My latest cartoon for @theguardian.com books. p.s. I have a new book of cartoons out soon: www.tomgauld.com/comic-books-v2

"Pat, why do you carry that ridiculous 600mm lens on long hikes?" Buddy, I can see mountains reflected in the eyes of a trailside pika.

NEW: After numerous tragedies, OpenAI says it's scanning users' conversations, escalating to human readers, and contacting law enforcement where necessary futurism.com/openai-scann...

🚀 gratia 0.11.0 is out! Now has a paper in JOSS — please cite 📄 doi.org/10.21105/jos... Experimental parallel processing ⚡ New assemble() for building plots 🎨 Better support for complex families + new diagnostics 🧪 Lots of bug fixes + polish ✨ 👉 gavinsimpson.github.io/gratia/ #Rstats

In case you didn't know: You can run your package unit tests with {testthat} in parallel with two simple steps (see testthat.r-lib.org/articles/par...): tldr: 1. Add `Config/testthat/parallel: true` to DESCRIPTION. 2. Add `TESTTHAT_CPUS=8` to your .Renviron and restart R. #rstats

The Dell XPS I bought in 2012 is my fav laptop EVER. Named 'rocinante', it features a 4 cores Intel i7, 7.3 GB RAM, 1TB SSD, and now runs Xubuntu 24.04, which makes it feel like 10 years younger. But its age is showing (I hear you, old buddy!), and it's no longer suitable for heavy-duty stuff.

Solid lesson! I’ve fallen into this trap myself: I was convinced my solution was right (when you are right, you are right, right?) and have never cared about workplace politics. This situation is facilitated by managers hiding context or playing tech lead without the required knowledge.

If there were a single lesson I'd drill into new software engineers it'd be this: You can be technically right and politically and systemically dead wrong.

Meanwhile, in Industry: - arithmetic operations? AI! - bunch of SQL queries? AI! - bunch of ifelses in a trenchcoat? AI! - tree-based model? AI! - k-means clustering? AI! Everything looks like AI if one is dumb enough.

Stretching DuckDB w/ Common Crawl, ~1.7B rows, ~300 parquet files. ~2-3s for single-column aggregations, ~2-3 mins to SUMMARIZE the data, peaking at ~12-14GB memory usage. Not exactly real-time, but the fact you can do this on a laptop with no server setups or Spark pipelines is still amazing.

Yup. If you have a reproducible example we can surely cook up a demo as posted over the years as gif/mp4 files. Eg 'tidyverse in 18 secs' or 'tidyverse + sf + rstan in 30-some secs'. Have a new (faster) machine so tidyverse is now 7s for me 🤩. #r2u: Fast. Easy. Reliable. Pick All Three. #rstats

I'm halfway through a 30-day unpaid leave. I took it to address work–life balance issues, care for my son, and reflect on my career (whatever that means). 30 days won't fix the challenges at home or work, but 💩 < 💩💩, which is enough to make it worth it.

If Ted Chiang, a truly superb creative writer (also; a background in compsci), is happy using the gym robot analogy to explain why using chatbots at university is antithetical to the purpose you're paying money and forgoing paid work to be there then I'm fortified in my view it's a good one.

Frustrating loop when installing certain #rstats packages in a fresh xubuntu: 1. install.packages("whatevs") 2. NOPE!, try installing these system libraries: ... 3. sudo apt-get install ... 4. back to 1 (again) I understand it's hard to achieve, but I wish R could resolve system deps by itself.

I watched eps 1 and 2 of #alienearth, and here's my useless take: - Production value: 🚀🚀🚀 (it looks AMAZING!) - Fan service: ✅ - PSYCHO alien bugs: 🐙🐙🐙 - Gore: 😱 - Cute kitty: 🐱 - Cartoonish techbro: 👎 - Unrealistic reactions: 👎👎👎 It did not meet my expectations, but it was still fun.

High dimensional spaces will not be flattened !

Clustering is harder than it looks!

Do you love alpine plants, ancient DNA, and/or population genetics? My lab will be hiring a postdoc soon to be based at the University of Maine! We’re looking for someone who uses genetic tools and thinks like a community ecologist. Happy to chat at #ESA2025 if you’re interested.

Seeing more posts like this (and I don’t mean to single out Andrea!) After enthusiasm thanks to rapid progress —> a few months of increasing pain —> giving up Turns out having a tool that generates code doesn’t remove the need to know how to program to build complex software…

My copy's in the post and should be arriving today! 😁

🗺️ ¿Qué nuevas perspectivas pueden surgir al trabajar con datos espaciales en R? Ya está abierta la inscripción para la nueva edición del curso "Mapas y Operaciones Espaciales (SIG) con R" https://www.transmittingscience.com/courses/ecology/mapas-y-analisis-espacial-sig-con-r/

There are some good hills to die on. This is one of them.

My attempt to stop drinking coffee lasted 36 hours and a threat of divorce, and I'm not even married #grumpyAF

git config --global branch.sort -committerdate Sorts git branch by the last commit. Game changer.

Observation: I met several truly exceptional people along my professional path. There is a personality trait that they necessarily have and demonstrate — they are very generous in sharing credit and recognizing one's input, always lifting people around them to new highs, motivating to grow & dare 🌱

Which paleoclimate dataset do you use for your research? New research by @elinerentier.bsky.social published in Frontiers of Biogeography reveals that your choice matters —a lot— especially in mountains! doi.org/10.21425/fob...

Found this quote in the post "Nobody Actually Cares" published at luminousmen.com (which is, by the way, an amazing data engineering blog!)

orbital is such a cool idea — convert your tidymodels or (now) scikit learn model predictions to SQL so you can run them in the database