iiinova — where your digital self keeps evolving

Reposted by Beeps

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…

15546205

Reposted by Beeps

Giles Wilkes @gilesyb.bsky.social · 26/04/2026

I have done the workings on whether Europe is doing as well as the USA after all, once you tot up all the social ills risked by the typical US citizen. open.substack.com/pub/gileswil...

open.substack.com

Maybe Europe is winning after all

(over) regulation is for a reason

1015451

Beeps @willbeeps.bsky.social · 25/04/2026

Just saw a man on the M25 using his knees to steer while playing an oboe. Just checked the Highway Code, I'm sure it's legit.

000

Reposted by Beeps

vibe-free lifestyle and smiling apple purveyor @ellaguro.bsky.social · 26/03/2026

my experience with the social media age has just been witnessing people failing to conceptualize that two ideas can be true at once over and over. there is a always rabid demand for one thing to be true to the exclusion of something else, and it never stops, and there is never any slowing it down.

35316

Reposted by Beeps

Tyler McBrien @tylermcbrien.com · 10/03/2026

The Singularity is upon us: Apple's AI summarized a text message that read "I'm still down to clown if you are" to.... "clown event still possible."

10896692019

Reposted by Beeps

Julia Carrie Wong @joolia.bsky.social · 07/03/2026

We don’t talk enough about how morally depraved the tech industry turned out to be. Every single ounce of their self-regarding statements of values was an outright lie.

12462581309

Beeps @willbeeps.bsky.social · 06/03/2026

Reading an AI-generated article on a crypto site telling me all the technical details of AI murder-drones and the venture capitalists that fund them. Time to go check on my seedlings to counteract this barrage of dystopia. www.mexc.com/en-GB/news/8...

mexc.com

Unveiling the first high-level assassination operation led by AI: How did Claude and Palantir kill Khamenei? | MEXC News

Source: Microsoft AI Written by: Xia Chu Today, a chilling stillness hangs over Shemiran, a district in northern Tehran. For Iran's Supreme Leader Ayatollah Ali

000

Beeps @willbeeps.bsky.social · 31/12/2025

As I saw our first proper frost of the year only this morning, I feel duty-bound to whinge that we're doing the calendar All Wrong. Come on, we've barely left autumn. We should have made the vernal equinox, or chocolate jesus day, the one with all the champagne and countdowns. Happy same year all!

010

Beeps @willbeeps.bsky.social · 27/09/2025

Sometimes a direct quote is so confronting you can't quite believe it: "I am directing Secretary of War to provide all necessary Troops to protect War ravaged Portland... I am also authorizing Full Force, if necessary." It's official, then: America is at war with itself. This is maniacal.

000

Beeps @willbeeps.bsky.social · 19/06/2025

"Thousands of UK university students caught cheating using AI". The language around AI in education is always of transgression, wrongdoing. Meanwhile, over on Snapchat...

000

Reposted by Beeps

Unwinder @unwinder.bsky.social · 10/05/2025

Generative AI has helped me to understand why, in Star Wars, the droids seem to have personalities but are generally bad at whatever they're supposed to be programmed to do, and everyone is tired of their shit and constantly tells them to shut up

3352671189

Beeps @willbeeps.bsky.social · 05/02/2025

The moon be out here looking like a Jean Michel Jarre album cover.

A lunar halo around the moon (obvs) in a clear night sky

000

Reposted by Beeps

Dr. Abigail Desmond @abigaildesmond.bsky.social · 13/01/2025

Ursula K. LeGuin on technology

11573701987

Beeps @willbeeps.bsky.social · 09/01/2025

Two perfect cat beans in the snow, placed a stretch away from a door. No further beans are in the picture.

020