I am excited beyond description to lift the veil on what we have been working on in 2026:
Please meet ggsql! A new extension of the SQL language for creating visualisations using the grammar of graphics. Read all about it in the blog post or visit the website at ggsql.org
I loved reading this. It’s so well written!
How to tell if a statistical method really works?
Monte Carlo simulations give a clear answer. Instead of relying on one dataset, you generate data repeatedly and compare methods across many runs.
New module in the Statistics Globe Hub: statisticsglobe.com/hub
#Statistics #DataScience #RStats
The semicolon thing could have come from copy/paste from a website that has the quote of interest. I've done that before too.
I've read of tons of AI-generated job applications, and I can probably predict those well.
My radar didn't go off when reading this (and I'm not here to fight you over it)
Genuinely, it's all circumstantial and IMO significant extrapolations.
I've read (and written) things that are not optimal, and the things that you attribute to AI don't strike me as being a "mistake that only an LLM would make."
The panache #quartopub formatter and linter by @jolars.co is excellent!
Very configurable and uses external tools (e.g., air or ruff) to format code blocks.
panache.bz
#quartopub version 2 is being cooked up in the lab.
@cscheid.net has a post discussing the why and when (with some hints about some related future projects)
quarto.org/docs/blog/po...
I didn't find this convincing in the least.
For me, this could not have been any more on target.
Still waiting…
Yes, and there are a few other python libraries that I'd love to wrap.
We've released the first version of our tabpfn #rstats package to CRAN. This is an interface to the Python #TabPFN package.
tidyverse.org/blog/2026/03...
The model is a pre-trained deep learning model that has performed exceedingly well on every data set I've tested it on.
I see what you did there 😀
Agreed.
The Kolmogorov-Smirnov statistic is one of those things like the ROC curve in that I absolutely understand how it’s created and what it means, but if somebody gives me a particular value of it, it’s very hard for me to conceptualize with the underlying curve(s) looked like.
I guess my beef with Shapley values is that I don't find them particularly intuitive; I'd rather see visualizations like partial dependency plots.
If a statistic is needed, basic permutation- or model-based importance scrores are more my thing (and are not perfect).
YES
The third date I had with my wife during Covid was her renting out a theater for $100. We saw Empire Strikes Back. 💯
Anyway, Project Hill Mary is very good. The acting and cinematography are perfect and it sticks pretty close to the book which was already amazing. 2.5h went by pretty quickly.
Not a good time for anyone at 10:30 pm on a Saturday.
Would. Not. Recommend.
Absolutely! He will defending his thesis next week so he’s about done with his presentation. Topic: “How My Brother Got Us Skunked Last Week”
Our list of 2026 #rstats and #python summer internships has been posted.
We can't wait to work with you and make great things!
tidyverse.org/blog/2026/03...
We’re thrilled to welcome Sara Altman and Simon Couch of the Posit AI Core team to the #positconf 2026 keynote stage!
They’ll be sharing a practical, hype-free look at how AI is being thoughtfully integrated into the tools you use every day.
✨ See what’s next for AI in open source: pos.it/conf
Cool little details about how to structure tree splits.
And don't worry, the cat is not in the dryer. I checked.
I'm over the moon excited to the release of 0.5.0 of orbital 🛰️
This release adds full support for boosted tree, faster creation of orbital objects, optimization of execution!
We can finally reliably predict with a xgboost model from a database!
tidyverse.org/blog/2026/03...
#rstats #tidymodels
something big is coming 🛰️
I’ve also made some headway on the python companion, but I’m not quite ready to release that yet.
Slow and steady! I’m hoping to have (single) classification tree chapter done in the next month and then probably off to boosting.
It’s hard to determine what a release should be since we’re going nonlinearly through the chapters (that pun was intended).
🤩
@drob.bsky.social’s talk from 7 years ago covers this well
youtu.be/dT5A0sAWc2I
I presented at Shiny in Production 2025, an incredible conference hosted by @jumpingrivers.com up in Newcastle! I was glad to share the very latest from the Shiny team directly. The topics were bleeding edge at the time, so still new and relevant now. My video: www.youtube.com/watch?v=vxai...
A pink and blue graphic reading "apply for our opportunity scholarship to posit::conf(2026)."
We are covering 40 people's travel, lodging, and registration for posit::conf() this fall! If you are from a group that is underrepresented in data science or open source, please consider applying for the Opportunity Scholarship—we'd love to have you join.
posit.co/blog/apply-t...
