iiinova — where your digital self keeps evolving

Steen Flammild Harsted @steenharsted.bsky.social · 03/06/2026

I really really like everything about the new Positron IDE... Except changing the settings. Can someone port the Rstudio settings pane to Positron please? #Rstats @posit.co

030

Steen Flammild Harsted @steenharsted.bsky.social · 03/06/2026

The internet has destroyed me... I was sure 'fwb' was short for Friends With Benefits

120

Steen Flammild Harsted @steenharsted.bsky.social · 21/04/2026

png() vs ggsave() Why does this code produce so different results? End goal is to have in-memory access to the rendered plot using unigd::ugd_render() #Rstats

130

Steen Flammild Harsted @steenharsted.bsky.social · 08/04/2026

Incoming game changer with Quarto 2.0 🥳 In 6 months I might never open a Word editor again ... well, at least not in projects where I am PI #rstats

141

Reposted by Steen Flammild Harsted

Cédric Scherer @cedricscherer.com · 12/12/2025

I ignored the strip.clip argument in #ggplot2 for way too long 😲 Combined with a small negative margin tweak, you can place facet labels inside each panel. A tiny trick that makes small multiples feel so much cleaner. 🔵 no manual coordinates 🔵 inherits theme styling 🔵 scales nicely when resizing

Side-by-side comparison of two multi-panel bubble charts faceted by world region. The left column shows the default facet labels placed above each panel (“Africa”, “Americas”, “Asia”, “Europe”, “Oceania”). The right column shows the same charts, but the facet labels are moved inside each panel at the top-left using a negative margin. In the center, there is a title reading “Want to place your facet labels inside each panel?” with an arrow pointing right, followed by a short ggplot2 theme code snippet demonstrating how to move strip text inside the panel.

724140

Steen Flammild Harsted @steenharsted.bsky.social · 29/10/2025

You can help Ukraine and learn R at the same time. What's not to like? Check out the workshops - seriously they look amazing

000

Steen Flammild Harsted @steenharsted.bsky.social · 20/10/2025

Am I wrong, or does Positron default to the same ‘I do not care about reproducibility’ settings as RStudio? For those who remember the good old Twitter days—here’s a throwback to a classic tip for settings in RStudio. Can we get a similar guide for Positron? @posit.co @hadley.nz #rstats

260

Steen Flammild Harsted @steenharsted.bsky.social · 08/10/2025

I just spent an entire day with ChatGPT-5 trying to fix a Shiny app that mysteriously crashed when deployed to our server. We tried everything. Nothing worked. We even broke the app worse. Then spent hours slowly resurrecting it. At 00:30 I suddenly remembered: Shiny Assistant by @posit.co

gallery.shinyapps.io

Shiny Assistant

130

Steen Flammild Harsted @steenharsted.bsky.social · 12/09/2025

In ellmer, is there any way to inspect the actual JSON schema that type_* creates? e.g. the opposite of type_from_schema() type_array( type_object( x = type_boolean(), y = type_string(), z = type_number())) |> type_to_schema() # <- what I need, but dont have @posit.co #rstats

020

Steen Flammild Harsted @steenharsted.bsky.social · 08/09/2025

That circle + angles is going to help me sooo much 🤩

010

Reposted by Steen Flammild Harsted

Julia M. Rohrer @dingdingpeng.the100.ci · 25/08/2025

Ever stared at a table of regression coefficients & wondered what you're doing with your life? Very excited to share this gentle introduction to another way of making sense of statistical models (w @vincentab.bsky.social) Preprint: doi.org/10.31234/osf... Website: j-rohrer.github.io/marginal-psy...

Models as Prediction Machines: How to Convert Confusing Coefficients into Clear Quantities

Abstract
Psychological researchers usually make sense of regression models by interpreting coefficient estimates directly. This works well enough for simple linear models, but is more challenging for more complex models with, for example, categorical variables, interactions, non-linearities, and hierarchical structures. Here, we introduce an alternative approach to making sense of statistical models. The central idea is to abstract away from the mechanics of estimation, and to treat models as “counterfactual prediction machines,” which are subsequently queried to estimate quantities and conduct tests that matter substantively. This workflow is model-agnostic; it can be applied in a consistent fashion to draw causal or descriptive inference from a wide range of models. We illustrate how to implement this workflow with the marginaleffects package, which supports over 100 different classes of models in R and Python, and present two worked examples. These examples show how the workflow can be applied across designs (e.g., observational study, randomized experiment) to answer different research questions (e.g., associations, causal effects, effect heterogeneity) while facing various challenges (e.g., controlling for confounders in a flexible manner, modelling ordinal outcomes, and interpreting non-linear models).

Figure illustrating model predictions. On the X-axis the predictor, annual gross income in Euro. On the Y-axis the outcome, predicted life satisfaction. A solid line marks the curve of predictions on which individual data points are marked as model-implied outcomes at incomes of interest. Comparing two such predictions gives us a comparison. We can also fit a tangent to the line of predictions, which illustrates the slope at any given point of the curve.

A figure illustrating various ways to include age as a predictor in a model. On the x-axis age (predictor), on the y-axis the outcome (model-implied importance of friends, including confidence intervals).

Illustrated are
1. age as a categorical predictor, resultings in the predictions bouncing around a lot with wide confidence intervals
2. age as a linear predictor, which forces a straight line through the data points that has a very tight confidence band and
3. age splines, which lies somewhere in between as it smoothly follows the data but has more uncertainty than the straight line.

461000285

Steen Flammild Harsted @steenharsted.bsky.social · 20/08/2025

What an awesome update. 100% going to use this when making slides for lectures.

010

Steen Flammild Harsted @steenharsted.bsky.social · 07/08/2025

Happy to say that 'mocapr' now supports import of joint center positions via .trc files from OpenCap. github.com/steenharsted... #rstats #mocap #opencap

041

Steen Flammild Harsted @steenharsted.bsky.social · 30/07/2025

Dear @posit.co, I'm thinking about switching to Positron, but I heard the `here` package only works with RStudio projects. I am afraid someone will set my computer on fire if I make the switch? Please advise before the flames start. 🔥💻🧯 #rstats

240

Steen Flammild Harsted @steenharsted.bsky.social · 26/07/2025

I love hos Posit seems to be working non-stop on making packages that integrate LLMs into R and data-science workflows

020

Steen Flammild Harsted @steenharsted.bsky.social · 20/07/2025

I am looking to build a speech-to-speech chat-bot in Shiny. Is this possible with the current ellmer and shiny-chat architecture? Thanks #rstats #shiny

110

Reposted by Steen Flammild Harsted

PoliticsGirl @politicsgirl.bsky.social · 11/06/2025

Yay AP called the NJ Governor Dem primary for our girl Mike Sherrill!!! youtube.com/shorts/iUFWG...

youtube.com

One More Badass!

YouTube video by PoliticsGirl

10688125

Steen Flammild Harsted @steenharsted.bsky.social · 05/06/2025

🤩patchwork 1.3.0 makes it possible to combine gt tables and plots 🤩 www.tidyverse.org/blog/2024/09... Thank you @thomasp85.com and @teunbrand.bsky.social #rstats

0144

Reposted by Steen Flammild Harsted

Hadley Wickham @hadley.nz · 03/06/2025

Want to easily use LLMs from R? ellmer 0.2.0 provides: • Batch & parallel processing • A universal parameter interface • Cost tracking • Support for Mistral, Hugging Face, & Portkey And much more! Read all about it at: www.tidyverse.org/blog/2025/05... #rstats

tidyverse.org

ellmer 0.2.0

ellmer 0.2.0 lands with a swag of upgrades: Garrick Aden‑Buie joins the team, we make a couple of breaking changes, and add serious scale with `parallel_chat()` and `batch_chat()`. A new `params()` ...

17526

Reposted by Steen Flammild Harsted

Stephen Turner @stephenturner.us · 02/06/2025

The Modern R Stack for Production AI: doi.org/10.59350/z9f... Python isn't the only game in town anymore: R can interact with local and cloud LLM APIs, inspect and modify your local R environment and files, implement RAG, computer vision, NLP, evals, & much more #Rstats

doi.org

The Modern R Stack for Production AI

Python isn't the only game in town anymore: R can interact with local and cloud LLM APIs, inspect and modify your local R environment and files, implement RAG, computer vision, NLP, evals, & much more

27117

Steen Flammild Harsted @steenharsted.bsky.social · 24/05/2025

How can I change the assisant logo/image/icon in shinychat? This is not working: chat_ui("chat", placeholder = "The patient is in front of you...", assistant_img = "logo.png" ) #rstats

230

Reposted by Steen Flammild Harsted

JazzReads📚🏀 @jazzreads.bsky.social · 10/05/2025

My love of reading is strongly link to this man…

354147383496

Reposted by Steen Flammild Harsted

Timothy Snyder @timothysnyder.bsky.social · 04/05/2025

"On Tyranny" is a #1 NYT bestseller again. I wish the moment were different. But I’m glad the book is useful. And grateful for all the kind words about putting the 20 lessons to work. snyder.substack.com/p/twenty-les...

snyder.substack.com

Twenty Lessons, read by John Lithgow

Key selections from On Tyranny, for viewing and sharing

375165284194

Reposted by Steen Flammild Harsted

Timothy Snyder @timothysnyder.bsky.social · 02/05/2025

It’s strange how we cannot make policy to protect Americans from Russian and Chinese disinformation without elements of the American far right identifying with and protecting Russian and Chinese disinformation. www.technologyreview.com/2025/05/01/1...

technologyreview.com

Senior State Department official sought internal communications with journalists, European officials, and Trump critics

Trump appointee Darren Beattie requested records regarding a large list of high-profile names, organizations, and right-wing buzzwords for a “Twitter files”-style document dump about alleged conservat...

361532475

Reposted by Steen Flammild Harsted

Mother Jones @motherjones.com · 28/04/2025

According to Kash Patel’s own financial disclosure statement, he pocketed $25,000 from a production company operated by a filmmaker associated with a Kremlin-subsidized propaganda project, a pro-Putin oligarch, and a pro-Kremlin disinformation agent. @davidcorn.bsky.social's NEW video explains:

youtu.be

The Curious Case of Kash Patel's $25,000

YouTube video by Mother Jones

499147615940

Reposted by Steen Flammild Harsted

Posit @posit.co · 21/04/2025

We’re thrilled to announce the release of rsample 1.3.0! rsample makes it easy to create resamples for assessing model performance. New updates include flexible grouping for bootstrap intervals and updates from tidyverse developer day. Learn more: www.tidyverse.org/blog/2025/04... #RStats

0311

Reposted by Steen Flammild Harsted

Chris Murphy @chrismurphyct.bsky.social · 17/04/2025

The Trump-Supreme Court battle is not really the crisis. The crisis is here now. Trump is enacting an insidious coordinated attack on our institutions of democratic accountability, designed to crater democracy before next fall. 1/ A long 🧵to explain the plan & how we stop it.

15545361929

Reposted by Steen Flammild Harsted

Hank Green @hankgreen.bsky.social · 21/04/2025

People spend a lot of time wondering why the left is worse at using independent media to control the national narrative. A useful exercise is to ask why the left is also worse at selling snake oil and pyramid schemes, because I think the reasons are basically exactly the same.

967428336853

Steen Flammild Harsted @steenharsted.bsky.social · 17/04/2025

What the past weeks have taught me about the U.S.: Americans are pretty meh about losing democracy, but god help you if you touch the economy. If Trump falls because of the economy and not for gutting democratic norms, I doubt the next, slightly more competent autocrat will be stopped at all.

000

Reposted by Steen Flammild Harsted

Posit @posit.co · 14/04/2025

We are thrilled to announce Air, an extremely fast R formatter! This super speedy tool written in #Rust ensures consistent, readable #RStats code. Read more on the tidyverse blog: www.tidyverse.org/blog/2025/02...

38417

Reposted by Steen Flammild Harsted

Dylan Pieper @dylanpieper.bsky.social · 12/03/2025

Exciting new dev feature allows you to batch LLM responses for structured data and ask the model to judge the response and refine the data in each chat instance. #rstats github.com/dylanpieper/...

github.com

GitHub - dylanpieper/hellmer: Batch Processing for Chat Models

Batch Processing for Chat Models. Contribute to dylanpieper/hellmer development by creating an account on GitHub.

042

Reposted by Steen Flammild Harsted

Posit @posit.co · 03/03/2025

We are delighted to announce Reticulate 1.41! You can now simply declare your dependencies using `py_require()`, and Reticulate will handle the rest. This seamless experience is powered by uv, an extremely fast #Python package manager written in Rust. Learn more: posit.co/blog/reticul... #RStats

012229

Reposted by Steen Flammild Harsted

easystats @easystats.github.io · 28/02/2025

One function per week, this time with `parameters::model_parameters()`. The function returns a comprehensive, consistent ("tidy") output for regression models and many other statistical procedures, including Bayesian and mixed models. #rstats #easystats easystats.github.io/parameters/r...

1203

Reposted by Steen Flammild Harsted

Posit @posit.co · 26/02/2025

We are delighted to announce the release of ellmer, an R package for accessing the power of LLMs in R! ellmer supports a variety of model providers. Chat, tool call, stream API calls, extract structured data, and more! Learn about it here: posit.co/blog/announc... #RStats #AI #LLM

ellmer hex, consisting of a colorful cartoon elephant

810128

Reposted by Steen Flammild Harsted

Dylan Pieper @dylanpieper.bsky.social · 12/02/2025

Batch process ellmer chats using the hellmer extension! Parallel processing makes it fast 🏃💨, state persistence makes it safe 🦺, and the ability to use ellmer’s tools and structured data extraction makes it powerful 🐘. github.com/dylanpieper/... #rstats #LLM #AI

392

Reposted by Steen Flammild Harsted

Mark Rieke @markjrieke.bsky.social · 13/01/2025

reaching out to the #rstats / #databs hivemind --- I'm looking for a maintainer! nplyr is a small R package I shipped to CRAN a few years ago that allows users to perform dplyr-like operations on nested data frames. If you'd like to take over, please follow up here: github.com/markjrieke/n...

03827

Reposted by Steen Flammild Harsted

Michel Nivard @michelnivard.bsky.social · 13/12/2024

Do I know any academic react/JS developers that would like to collaborate on an open source tool that’s to become “google docs for quarto/ipynb”? I might get it to work but I am a little out of my depth and a terrible UI designer..

241

Steen Flammild Harsted @steenharsted.bsky.social · 04/12/2024

Need help deploying a Shiny app with shinychat to shinyapps.io! Getting dependency errors with elmer package. Error is: 'unable to satisfy package: elmer (0.0.0.9000)'. Already tried explicit remotes in DESCRIPTION and install_packages.R. Any tips? #rstats #shiny"

shinyapps.io

210

Steen Flammild Harsted @steenharsted.bsky.social · 03/12/2024

Imagine Hadley Wickham has put laid the tidyverse to rest to focus more on grant writing... 😱

000

Reposted by Steen Flammild Harsted

Hadley Wickham @hadley.nz · 25/11/2024

I'm writing up a brief history of the tidyverse? What do you want to know about it? #rstats

83521111

Reposted by Steen Flammild Harsted

Khoa @khoavuumn.bsky.social · 01/12/2024

Reshape in Stata without looking at the help file

1037235

Steen Flammild Harsted @steenharsted.bsky.social · 03/12/2024

This is super useful! Never heard of the .keep argument before. How long has this been around?

010

Reposted by Steen Flammild Harsted

Prof. Emily Nordmann @emilynordmann.bsky.social · 02/12/2024

If you've ever wanted to learn #rstats then consider checking out our course Applied Data Skills which focuses just on data wrangling and viz and working in R rather than stats and will take you from zero to hero in reproducible reports. You even get walkthrough vids of my ramblings #AcademicSky

14612