iiinova — where your digital self keeps evolving

Roar Stovner @roarstovner.bsky.social · 07/10/2026

Simine Vazire living up to expectations as editor of Psych Science.

Reposted by Roar Stovner

Elizabeth Wrigley-Field @wrigleyfield.bsky.social · 16/09/2026

This stands out to me because I so often see academics give the opposite advice—save that for the next article or next book. It’s true that each piece of writing needs to be focused right to do its own work and not have readers lose their path on the journey you want to take them on. But in all the

1184

Reposted by Roar Stovner

Marc Veldhoen @marcveld.bsky.social · 07/09/2026

Switzerland's Federal Government is Replacing Microsoft on 3,000 Computers This is the pilot phase, with 3,000 workstations moving away from Microsoft 365. The federal administration has over 54,000 computers. itsfoss.com/news/swi... 1/2

itsfoss.com

Switzerland's Federal Government is Replacing Microsoft on 3,000 Computers

This is the pilot phase, with 3,000 workstations moving away from Microsoft 365. The federal administration has over 54,000 computers.

22414

Reposted by Roar Stovner

Ketan Joshi @ketanjoshi.co · 17/08/2026

This data centre features a 9.2 gigawatt fossil fuel power station Annually it'll pump out around 30 MTCO2-e. Same as - 8 coal plants - 9.2 million cars - Destroying 20 million hectares of forest - Half a percent of America's total national emissions drawdown.org/carbon-in-co...

Grid Investment
High Voltage Transmission
76SkV Transmission Lines
$4.2 Billion
Grid Investment
New Power
10 OW capacity
Including 9.2 OW of
Natural Gas Generation
as part of the U.S.-Japan Strategk: Trade aru3
Investrnent Agrement
$33 Billion
Gas Generation Invesment

a graphic showing the equivaelent emissions for various compariosn

7166117

Reposted by Roar Stovner

vincentholst.bsky.social @vincentholst.bsky.social · 13/08/2026

In 2023, @nature.com published 'Papers and patents are becoming less disruptive over time', receiving world-wide media attention. Our Matters Arising, published after a 32 month delay (more on that soon), shows that the reported decline can largely be attributed to dataset artefacts. 🧵

The average CD_5 index per year for Web of Science. The original Park et al. decline (top curve) becomes essentially flat (bottom curve) when removing papers with CD_5=1. Those papers largely correspond to dataset artefacts.

9352173

Reposted by Roar Stovner

Danielle Navarro @djnavarro.net · 06/08/2026

The title is a little pointed, but I am very serious here. Academics, medical researchers, and other organisations should not collect data on birth sex. It is rarely relevant, and it almost always leads to privacy violations for transgender people. Please share. I would like people to read this one.

blog.djnavarro.net

Separating sex and gender is statistical malpractice – Notes from a data witch

Statisticians and academics in Australia should ignore the terrible advice of most LGBTIQA+ organisations and read this post instead. The 2020 ABS standard is almost never the correct way to collect d...

14236119

Reposted by Roar Stovner

M.J. Crockett @mjcrockett.bsky.social · 31/07/2026

This is more than the NSF social sciences division - now on the chopping block - awarded in grants last year. Think about that: a single AI company has the money to fund more social science research than the US govt.

614077

Reposted by Roar Stovner

Ryan Moulton @moultano.bsky.social · 18/07/2026

My current intuitive understanding of distillation which may have been totally falsified in papers I haven't read (and I haven't read much.):

1113

Reposted by Roar Stovner

Ben Williamson @benpatrickwill.bsky.social · 18/07/2026

The anatomy of an AI-supported academic article is becoming more apparent. I desk reviewed ~70 manuscripts this week for our journal. Reckon 20ish had significant AI assistance. On top of hundreds of others recently I'm starting to see patterns. Are other editors are seeing the same or similar? 1/

28932418

Reposted by Roar Stovner

Robert Gieseke (Openclimatedata) @openclimatedata.net · 13/07/2026

If you haven't seen this yet, this chapter might be of interest. bsky.app/profile/njr0...

121

Reposted by Roar Stovner

Robert Reich @rbreich.bsky.social · 12/07/2026

Trump approved disaster aid requests from six red states last week. Two days later, he rejected aid requests from four blue states. Not only is he punishing states who didn’t vote for him, he’s playing politics with people’s lives. Despicable.

35166182941

Reposted by Roar Stovner

Göktuğ Kayaalp @igoktugk.bsky.social · 04/07/2026

Let me do a little thread on writing software. I'm no expert, just a curious nerd who's tried many.

44520

Reposted by Roar Stovner

Mickaël CANOUIL, Ph.D. @mickael.canouil.fr · 21/06/2026

Every line, star, and gradient in this chart was drawn in Typst, with Gribouille plotting its own star history. Achieve publication-quality graphics with a single toolchain, no external plotting tools needed. Gribouille 0.4.0 soon. Full source ships with the announcement. #Typst #DataViz

Midnight-sky line chart of Gribouille's cumulative GitHub stars per day from April to June 2026, titled "Gribouille, a Rising Star". A luminous gold step trail rises from zero across a long flat run labelled "Quietly built in private", lifts at a gold dashed marker labelled "Made public 17th of May", then climbs through dashed release markers (v0.1.0 onward) to a final amber spike of 178 stars, annotated "+99 in a day". The y axis counts stars in steps of 25; the x axis spans months.

3152

Reposted by Roar Stovner

Noam Ross @noamross.net · 17/03/2026

I must complete the required security training. To do so, I must use the learning platform. The learning platform requires signing up for the new credential system. There is a training to sign up for the new credential system.

1314128

Reposted by Roar Stovner

Voilà: Francis Gagnon @chezvoila.com · 08/06/2026

Colour is so complex and yet it's often the first thing that people want to use to encode information. I see it as a last resort and with limited use. 📊

1151

Roar Stovner @roarstovner.bsky.social · 02/06/2026

Ben Recht's blog is always worth reading, at least if you like historical essays about science with epistemological implications.

010

Reposted by Roar Stovner

Sholto David @sholtodavid.bsky.social · 02/06/2026

Today I'm posting a new blog about an astonishing scientific own goal. Hundreds of papers have reported using a totally wrong antibody to investigate the tumor suppressor p16. This mistake has happened because scientists have muddled the names of two proteins. forbetterscience.com/2026/06/02/m...

1617078

Reposted by Roar Stovner

Jason Koebler @jasonkoebler.bsky.social · 01/06/2026

New: Hackers have been stealing high-profile Instagram accounts by simply asking Meta's AI support chatbot to change the email associated with the account they want to steal. Shockingly easy, terrible flaw associated with offloading support to AI: www.404media.co/hackers-simp...

404media.co

Hackers Simply Asked Meta AI to Give Them Access to High-Profile Instagram Accounts. It Worked

The exploit shows the extreme risk of offloading technical support to AI.

431790858

Reposted by Roar Stovner

em @foremphasis.bsky.social · 01/06/2026

yep

512520

Reposted by Roar Stovner

Nick Brown @steamtraen.eu · 29/05/2026

This is a really interesting thread. As someone who wrote his first computer program in 1975, my view is: If you have some cool code, just give it away as-is, CC-0. Only make a formal package (in whatever ecosystem) if you are prepared to commit to it like a child, including 18 years of support.

3112

Reposted by Roar Stovner

John Sakaluk @johnsakaluk.bsky.social · 28/05/2026

A small 🧵 about frustrations and opportunities w/ open-source research software for the social sciences: Came across a new paper in a top methods journal, explaining/demo'ing a new R package. Sounded v. cool, so was excited to dig in. What I found was largely a mess...

43210

Reposted by Roar Stovner

Sasha Gusev @sashagusevposts.bsky.social · 23/05/2026

I assigned random gender/ethnicity labels to scientific abstracts from the literature and then asked Claude to do a thematic analysis. Claude identified a clinical versus computational split for female/male authors and a DEI focus for Black/URM authors. All in completely random data.

26865340

Reposted by Roar Stovner

Philip N Cohen @philipncohen.com · 22/05/2026

My family always listened to CBS World News Roundup at 6pm in the 1970s. Then I was a reporter for CBS Ithaca affiliate WHCU in 1985-87. Pay and status were below print and TV, but radio was cool, and CBS was the most serious network. Tonight CBS radio ends service (to 700 stations) after 99 years.

Me at the news desk in the WHCU offices, 1985-7.

510237

Reposted by Roar Stovner

Chris Kenny, PhD @chriskenny.bsky.social · 21/05/2026

If you do record linkage in #rstats, I'd love your feedback on irelink: Fast Probabilistic Record Linkage. It is a port of Python's splink package, rewritten with a clean, pipeable R interface. All of the magic happens in a database backend (default is DuckDB) so it's incredibly fast.

Find duplicate records within a single dataset. The bundled fake_20 dataset provides a small example for experimentation. irelink pushes data into a SQL database for efficient pair generation, so you need a DBI connection. Here we use an in-memory DuckDB instance.

An il_spec describes the linkage model: which fields to compare, how to compare them, and which blocking rules to apply. Blocking rules restrict which record pairs are generated. Only pairs that share a surname or first name are scored, keeping computation manageable. il_estimate_u() estimates parameters for non-matching pairs via random sampling, and il_estimate_em() refines the match-weight parameters using Expectation-Maximization.
spec <- il_spec() |>
il_compare(first_name, cl_jaro_winkler(0.9, 0.7)) |>
il_compare(surname, cl_jaro_winkler(0.9, 0.7)) |>
il_compare(dob, cl_exact()) |>
il_block_on(surname) |>
il_block_on(first_name)

model <- il_model(df, spec = spec, con = con)
model <- il_estimate_u(model)
model <- il_estimate_em(model, block_on(surname))

predict() scores all candidate pairs and returns those above the match-probability threshold. The returned match_weight is the evidence-only log2 Bayes factor; total_match_weight adds the prior odds used to compute match_probability. il_cluster() groups the matched pairs into deduplicated entities.

2174

Reposted by Roar Stovner

Per Engzell @pengzell.bsky.social · 20/05/2026

Good teachers don't just raise the achievement of students they teach. They also shape the teachers of tomorrow. Standard estimates of teacher impact ignore these intergenerational spillovers. Thread on new work led by @danilok.bsky.social. Paper: osf.io/preprints/so... 1/10

15814

Reposted by Roar Stovner

Per Engzell @pengzell.bsky.social · 11/05/2026

New collider bias example just dropped

INSTITUTE FOR PERFORMANCE NEUROLOGY: Among elite young chess players, the kids with lower IQs were the stronger players. Chess is considered one of the most "intellectual" games in the world. But IQ didn't predict who was good at it. Swipe to see the research.

812918

Reposted by Roar Stovner

Posit @posit.co · 11/05/2026

Quarto 1.9 brings Typst even closer to LaTeX parity with 📖 Typst Books, ✍️ margins for figures, tables, and citations via the Marginalia package, and 🛜 new typst-gather tool for reliable, offline rendering. Details here: quarto.org/docs/blog/po...

08419

Reposted by Roar Stovner

Ai2 @ai2.bsky.social · 05/05/2026

Robotics models often struggle outside controlled environments. Ours is built to work in real ones. Today we're launching MolmoAct 2, which can assist with a host of chores & lab tasks, plus the MolmoAct 2-Bimanual YAM dataset—the largest open robotics dataset of its kind. 🧵

26918

Reposted by Roar Stovner

Berna Devezer @devezer.bsky.social · 30/04/2026

📣 The difference between replicable and not replicable is not itself scientifically replicable. 📣 New work with Erkan Buzbas, showing that verdicts such as "X% of results replicated" are based on an inferential machinery that doesn't work. arxiv.org/abs/2604.26268

Screenshot of a paper's title page. Title: "The Difference Between 'Replicable' and 'Not replicable' is not Itself Scientifically Replicable". Authors: Berna Devezer and Erkan O. Buzbas, both at the University of Idaho — Devezer in the Department of Business and the Institute for Modeling Collaboration and Innovation, Buzbas in the Department of Mathematics and Statistical Science. Authors contributed equally. Corresponding author: bdevezer@uidaho.edu.
Abstract: Replication studies estimate the replicability rate of scientific results by aggregating binary verdicts of experiments. Exact replications are rarely attainable, so most replication sequences are non-exact. Experiments differ in ways that matter and do not share a single common data-generating process. We formalize two statistical interpretations of this non-exactness. In a shared latent rate model (benchmark), experiments are exchangeable and depend on a common random replicability rate. In a conditionally independent rates model (operational), each experiment has its own replicability rate drawn independently from a population distribution. Under the shared latent rate model, even small variability among replicability rates induces an irreducible variance floor on the estimated mean replicability rate that cannot be eliminated by adding more replications. Under the conditionally independent rates model, the degree of non-exactness is not identifiable from standard replication data, because one binary verdict per experiment contains no information about between-experiment heterogeneity. Researchers therefore cannot tell which precision regime they are operating in or whether high- and low-replicability sequences can be distinguished in principle. As a result, the usual data structure of one binary verdict per experiment cannot support reliable demarcation between "replicable" and "not replicable" results and systematically understates uncertainty, making high- and low-replicability sequences appear discrim…

15545205

Reposted by Roar Stovner

Ben Williamson @benpatrickwill.bsky.social · 24/04/2026

From journal editing experience this tracks. In 2022 our journal got ~300 submissions. In 2025? 1300. So far this year? 500 and we're not even out of April yet. The AI trickle is now a stream, yet T&F are *still* seeking new big tech partners to flog papers for AI training rather than *helping us*

17933

Reposted by Roar Stovner

Emil Hvitfeldt @emilhvitfeldt.bsky.social · 19/04/2026

If you are doing any #quarto slidecrafting, i have been trying really hard to keep this page updated with everything I have done emilhvitfeldt.com/project/slid...

Screenshot of grid of some of my quarto extensions in a grid layout

27221

Reposted by Roar Stovner

Vincent Arel-Bundock @vincentab.bsky.social · 17/04/2026

The great @eddelbuettel.com invited me to his STAT447 class at the University of Illinois. If you'd like to hear me speak about the interpretation of statistical models in #RStats, using the {marginaleffects} 📦, check out the video! www.youtube.com/watch?v=v3TX...

youtube.com

STAT 447 (2026) Guest Lecture by Vincent Arel-Bundock

YouTube video by Dirk Eddelbuettel

07019

Reposted by Roar Stovner

Tom Hardwicke @tomhardwicke.bsky.social · 11/03/2026

Has transparency improved in clinical psychology? New cross sectional study from us (Bianca Kotoulas, Justine Blackwell & me). Adoption of transparent research practices has increased between 2012, 2018, & 2024, but aside from coi and funding disclosures, they remain uncommon osf.io/preprints/ps...

Graph showing the proportion of clinical psychology studies that adopted four transparent research practices (preregistration and sharing of measurement instruments, data, and analysis scripts) at three time points (2012, 2018, 2024). The graph suggests adoption of these practices has increased but remains uncommon.

Graph showing the proportion of clinical psychology studies that adopted three transparent research practices (reporting guidelines and disclosure of conflicts of interest and funding) at three time points (2012, 2018, 2024). The graph suggests use of reporting guidelines has increased but remains uncommon. Conflict of interest disclosure statements have increased and are now quite common. Funding disclosure statements were relatively common across the three time points.

05921

Reposted by Roar Stovner

Miles McBain @milesmcbain.bsky.social · 10/04/2026

Guy I’m sponsoring very quietly releases #rstats package I’ve been wanting for the last couple of years. pkg.yihui.org/gglite/doc/g... Guess he shut up and took my money. 🤣🤣🤣

pkg.yihui.org

Introduction to gglite

A visualization in gglite is built by composing independent layers:

39821

Reposted by Roar Stovner

Carlos Scheidegger @cscheid.net · 06/04/2026

And another Quarto announcement; I've alluded to it before, but we're making it "official". We've started work on Quarto 2. The blog post has an overview: quarto.org/docs/blog/po... We'll share more in future blog posts, but here's what you can expect from the Quarto 2 dev effort: (1/)

quarto.org

What’s next: Quarto 2 – Quarto

We’ve started working on quarto-dev/q2, a full rewrite of Quarto in Rust.

1019254

Reposted by Roar Stovner

jonny (leachate composite form) @jo.nny.rip · 06/04/2026

Not sure how great it is for the state of science where rather than publishing code and working on shared libraries that make it so everyone can easily do exactly the same analysis, everyone just generates their own code privately that purports to do something and kinda looks right but not sure

3268

Reposted by Roar Stovner

Katie Corker @katiecorker.bsky.social · 26/03/2026

Open Research Europe is re-launching with a new platform, and broader set of supporters, this fall. It uses the publish-review-curate model. "Publishing will remain completely free for both European Commission-funded researchers and authors from participating countries." home.cern/news/news/ce...

home.cern

CERN to host Europe’s flagship open access publishing platform

In an important step for open science, CERN has been selected to host a new phase of Open Research Europe (ORE), an initiative supported by the European Commission and a new funding consortium of Euro...

1711

Reposted by Roar Stovner

Conrad Hackett @conradhackett.bsky.social · 24/03/2026

The flagship journal of the European Sociological Association, European Societies, makes published articles free for everyone & does not charge authors a publication fee. Authors of quantitative work are required to openly share data & code. No @asanews.bsky.social journal shares all these traits.

direct.mit.edu

Editorial: What we promised, and what we delivered

As we write our final editorial for European Societies, we return to the spirit of our first editorial (Präg, Ersanilli and Gugushvili, 2022) and to the question that guided our tenure: did we deliver...

16224

Roar Stovner @roarstovner.bsky.social · 23/03/2026

Reading Aarons posts teach me how little I understand about searching the academic literature. That's a nice thing to learn.

100

Reposted by Roar Stovner

Nick Holmes @thehandlab.bsky.social · 23/03/2026

One thing that really surprises me as #Reviewer & #Editor is that, given the same set of data, most researchers seem to prefer reporting a weak, confounded, positive result instead of a stronger, well-controlled, negative result. It's a lack of bravery &/or self-awareness &/or something else? 🧪🧠⚡

1102