iiinova — where your digital self keeps evolving

I can't believe Hytale is doing this

What @hytale.bsky.social is doing, especially around creators and modding, is a model for the #atmosphere #atdev @atproto.com www.youtube.com/watch?v=H3RA...

youtube.com

YouTube video by Kweebec Corner

220

Dear #UK, I propose a trade deal #NoKings for #NoConfidence

000

Interesting observation while starting my fresh backfill... Many PDSes have some form of "bluesky" in the available domains, as opposed to "atproto" #atdev

verdverm @verdverm.com · 10/03/2026

PLC op count pushing 100M atmunge=# select count(*) from plc_log_entries; count ---------- 85758128 I assume this means we are now over 50% invalid ops. #DataSci will give the answer soon enough

140

Some nice graphs and stats in this one

Jimmy Hoover @jimmyhoover.bsky.social · 10/03/2026

Reposted by verdverm

My exclusive story on the new program, featuring comments from federal public defender Geremy Kamens of Virginia's eastern district, who spearheaded it. "Every time we’ve had a Supreme Court case, we’ve sort of had to reinvent the wheel," said Kamens. www.law.com/nationallawj...

law.com

US Judiciary Approves New Public Defender Office to Improve Supreme Court Advocacy | Law.com

The federal judiciary on Tuesday approved the creation of an office dedicated to improving the advocacy of federal public defenders at the U.S. Supreme Court.

214725

not to distract from the real talk, but I would love to see an OpenClaw version of this

Pulling out of Iran is like pulling out during a one night stand... It was a mistake from the start and it will be really expensive if you don't

100

HN's thoughts on the big news, have your grains of salt handy, but also important to know what people outside the #atmosphere think too :] news.ycombinator.com/item?id=4731...

news.ycombinator.com

Foreign Affairs @foreignaffairs.com · 08/03/2026

Court filings for Anthropic v Government They are just sueing anybody it seems :] www.courtlistener.com/docket/72379...

000

Reposted by verdverm

The editors at Foreign Affairs have compiled some of the magazine’s best coverage of the war in Iran—and what might come next. Explore the list here: fam.ag/4cyAjCB

173

verdverm @verdverm.com · 06/03/2026

blog.system76.com/post/system7...

The challenges we face are neither technical nor legal. The only solution is to educate our children about life with digital abundance.

Laurens @laurenshof.online · 05/03/2026

Reposted by verdverm

There's also a related queston for the atmosphere here: if a news website whats to add a share to Bluesky button to their website, should that button say Bluesky? If yes, should there also be a button for Blacksky? If no, what should it say instead?

8121

verdverm @verdverm.com · 05/03/2026

@simonwillison.net reading through your clean room post, have you thoughts on the Ai output is not copyrightable, ergo it cannot be licensed either? SCOTUS refused to hear appeals a few days ago, so lower court ruling stands for now. This post explores, not a lawyer too tuananh.net/2026/03/05/r...

tuananh.net

Relicensing with AI-assisted rewrite

Exploring the chardet v7.0.0 controversy: Can an AI rewrite legally 'launder' a library from LGPL to MIT?

140

verdverm @verdverm.com · 05/03/2026

My choice clips, this piece is really good

The Doom Loop and Getting Out of the Race, title + full section text

Building With Clay but No Kiln, title + two paragraphs

Pekka Lund @pekka.bsky.social · 04/03/2026

Reposted by verdverm

Hinton is one of those rare people who truly understands how both AIs and humans work, and has also read philosophy from the right sources (he refer to Dennett there). And he's not shy about pointing out how most people have a hopelessly wrong view of how people work, which is the main issue.

121

Reposted by verdverm

dan sinker @dansinker.com · 04/03/2026

I've now printed and shipped over 20,000 whistles, sent across the country to groups that need em. That's 20,000 off a single 3d printer that has basically run continuously in my basement since about mid-January. The crew I print with has sent out over 1M now, and we can send to you too. toot.supply

939461

Reposted by verdverm

Kimberly - You can call me MS. "Fucking Bitch" @klr-reno.bsky.social · 04/03/2026

There are 350 North Atlantic Right Whales left. Their primary method of death is ship strikes. Since these protections were implemented in 2020 the population grew by around 5% - 7%. This rollback will be catastrophic.

savannahnow.com

NOAA plans to roll back protections for endangered right whales

NOAA aims to undo vessel-speed regulation that protects endangered North Atlantic right whales, instead looking at real-time detection alternatives.

33827

verdverm @verdverm.com · 04/03/2026

Open Renewable Energy?! open source ideals moving towards ubiquity is good for everyone* libre.solar

libre.solar

The Libre Solar Project

Building Blocks for DC Energy Systems

040

Reposted by verdverm

Alexander Kustov @akoustov.bsky.social · 04/03/2026

Can I just commend @jatucker.bsky.social and @solmg.bsky.social for writing a much better version of my viral argument for academics to wake up on agentic AI—with concrete use examples, no unnecessary provocation or AI-generated text, and a more constructive vision for the future. Read it!

3123

Reposted by verdverm

Henry @henrythedog.bsky.social · 03/03/2026

this is what awaits us if we fail to crush Platner in the primary, "oh the party of Nazi Tattoos has something to say?" every time forever

Tweet by Igor Bobic

"Fetterman responds to Graham Platner saying he should lose his seat:

“Oh, really? The guy with the Nazi tattoo? Yeah, I'm devastated by that.” "

852808539

A Rational Analysis of the Effects of Sycophantic AI subtext, being careful with what you ask for is not enough arxiv.org/abs/2602.14270

Analyzing How Sycophancy Distorts Beliefs

We propose sycophancy leads to less discovery and overconfidence through a simple mechanism: When AI systems generate responses that tend toward agreement, they sample examples that coincide with users’ stated hypotheses rather than
from the true distribution of possibilities. If users treat this
biased sample as new evidence, each subsequent example increases confidence, even though the examples provide no new
information about reality. Critically, this account requires no
confirmation bias or motivated reasoning on the user’s part.
A rational Bayesian reasoner will be misled if they assume
the AI is sampling from the true distribution when it is not.
This insight distinguishes our mechanism from the existing
literature on humans’ tendency to seek confirming evidence;
sycophantic AI can distort belief through its sampling strategy, independent of users’ bias. We formalize this mechanism
and test it experimentally using a rule discovery task.

021

Thought of two Portmanteaus today "hypster" = hipster + hype (cycler), maybe we can get "ai" into this one? "dopamatic" comment = replying on autopilot for the dopamine "portmantwo" = a bad self-recursive joke "portmantroi" = contextual collapse from being left to my own devices for too long

This is an excellent ai "generated" thought piece. "Generated" is in quotes because the author used his prior writing as part of the inputs. The results speak for themselves

Stefan Schubert @stefanschubert.bsky.social · 03/03/2026

Reposted by verdverm

In today’s newsletter: the immense power of negativity bias. It leads us to exaggerate social and environmental problems and to underestimate Western power. Correcting for this bias is central to a correct worldview. www.update.news/p/when-the-e...

update.news

When the environment recovers, no one talks about it

Plus: Britain’s inactivity crisis is receding, Western power is underestimated, and more

1155