flancian

Wait, did I get this right that the answer after all the investigation that showed they had set up a goblin-reinforcing loop during fine tuning was... to ask it to not mention goblins so much in the system prompt?!

"Not dead. Not retiring. Not moving on to start a blockchain company. He was quitting — because of a fight over two characters"

Time to AI tell: 20 seconds? Still, interesting account of how Guido stepped down.

Also, what happened afterwards, which I wasn't aware of:

  > Guido’s departure left Python without a governance model for the first time in its history. What followed was genuinely impressive: the core developer community proposed, debated, and voted on no fewer than seven different governance PEPs (PEP 8010 through PEP 8016).
  > In December 2018, PEP 8016 — “The Steering Council Model” — won. Authored by Nathaniel J. Smith and Donald Stufft, it established a five-person steering council elected by core developers. The design philosophy was explicit: “Be boring. We’re not experts in governance, and we don’t think Python is a good place to experiment with new and untried governance models.”

Very cool, thank you! Both the visualization tool and the description of Miller-Rabin.

I didn't know an algorithm with these properties existed!

Furthermore, your tool gave me a more intuitive feel of the rate at which primes "thin out" than every treatment of the topic I read previously.

It's federated as in ActivityPub.

I was first exposed to Popper in my first year in university, in an introductory course to epistemology, and I liked him right away. I've carried Refutability and the Paradox of Tolerance with me ever since.

By the way David Deutsch could be reasonably said to be Popper's biggest fan. If you're interested in Deutsch and Popper you could do worse than picking up "The Beginning of Infinity" to get acquainted with both, it's a great book.

I'd like to buck the apparent trend of reacting to your project with shock and horror and instead say I believe it's a great idea, and I appreciate what you are doing! People have been trained to believe (very long) copyright terms are almost a natural law that can't be broken or challenged (if you are an individual; other rules might apply to corporations...) but I think we are better off continuing to challenge this assumption.

I could imagine adding support for further rules that determine when Levin actively runs -- i.e. only run if the country or connection you are in makes this 'safe' according to some crowdsourced criteria? This would also serve to communicate the relative dangers of running this tool in different jurisdictions.

Found this archive link:

https://archive.is/2026.02.02-090119/https://www.wired.com/s...

File under "unexpectedly interesting and cool". I'd never seen such good visualizations of magnetic fields before, or of the magnetization process.

  I had downloaded the final Aldus TIFF specifications document, hoping to find the author’s name. However, the name is seemingly written in white text on white paper - making it invisible. What?

Is there an explanation for this that I missed? Was it an Easter Egg left by the author?

I believe this is likely true and will become evident in time to most, but not all, carbon based systems.

Immich is supposed to solve this nowadays: https://github.com/immich-app/immich

TLDR: a user of ChatGPT and weed became convinced they had chanced upon earth shattering discoveries while interacting with the chatbot over weeks/months.

About the person: https://en.wikipedia.org/wiki/Vera_Rubin

The observatory: https://en.wikipedia.org/wiki/Vera_C._Rubin_Observatory

No cred gained here :) I was just surprised by the low quality of the article and there were no other comments by the time I got here so I thought I'd write this short critique to warn others, but I think it came across as a shallow dismissal so it didn't add much.

Pretty weird article. Just on the title alone:

- 'Mysterious': there's very little of substance in the article to advance the position that it is mysterious at all.

- 'Sacred': OK, I guess!

- 'Pyramid': "A fourth explanation — one for which there is no evidence — is that Cerro El Cono sits on the ruins of a pyramid built by ancient Indigenous tribes". So... no evidence then :)

- 'Hidden': "It rises steeply from the relatively flat jungle landscape of eastern Peru, making it visible from as far west as the Andes — 250 miles (400 kilometers) away — on a clear day."

Hello fellow hobby factorizer :)

I wanted to share this small gem of a paper in case it helps you like it's helped me:

"Simple divisibility rules for the 1st 1000 prime numbers": https://arxiv.org/pdf/math/0001012

This is by Hans Moravec, of https://en.wikipedia.org/wiki/Moravec%27s_paradox fame; a student of John McCarthy.

Not the author but: yes. This word emerged from online discourse a few years back about 'wordcels' vs 'shape rotators':

https://en.wiktionary.org/wiki/wordcel

https://en.wiktionary.org/wiki/shape_rotator

I respectfully disagree, I think it makes sense to stress the distinction. Along the lines of: let's reclaim 'algorithm' from the entities that made the word into a bad thing for so many.

This is precisely https://social.coop, of which I am part of, except for the "not being Mastodon" bit :)

Governance happens on Loomio, which is a forum tailored for community decision making, but the main platform that coop members get access to is a Mastodon instance for now. We are experimenting with Bonfire, which is more flexible and might end up providing a solid base for federated apps in general.

Let me know if you have any questions!

Claude seems promising, but after giving it a try two things stood out to me (I have previously used ChatGPT and Gemini):

1. I miss voice I/O more than I thought I would, in particular on mobile. Maybe because of how great of a job Whisper has been doing in the case of ChatGPT in particular.

2. Claude seemed hyper-conservative in ways that I would personally classify as annoying. Example: it refused to detail the plot of a ~60 year old book, claiming that doing so would infringe copyright (!). It relented only after I pointed out you can find the same information on Wikipedia.

It has ActivityPub support, which in some ways (at least for me) is even better -- this means that you can e.g. follow Betula accounts from Mastodon (or Bonfire, etc.) and see what they bookmark.

What if it's garbage collection and NN training/backprop time?

Then it starts sounding like a great deal to me at least.

I've been an early access user for a few months and I'm really enjoying Beeper -- I hope they continue to deliver like this and can scale the operations as needed.

I currently use through it: Telegram, WhatsApp, RCS (Android Messages), Signal.

Beeper is built on Matrix and their client is based on Element but I would say the experience is slightly better than native; for example I prefer how client verification works in Beeper over vanilla Element.

Congrats on the open beta!

For others who were also stumped here for a bit: it seems your text got mangled, likely by triggering markup rules. ITYM 44100 = (2 x 3 x 5 x 7)^2

Even worse, I can confirm that logging in is not sufficient, despite what the initial call to action seems to claim.

After logging in the article is still incomplete and the dialog is changed to one prompting the user to "upgrade", meaning payment.

This seems like a dark pattern to me.

I think it's an opinion, not objective fact. I respect your opinion but as a fellow argentinian I don't think this is true in every case, or even in most cases.

Not the parent but I really enjoyed Scott Aaronson's "Quantum Computing Since Democritus":

https://en.wikipedia.org/wiki/Quantum_Computing_Since_Democr...

Based on:

https://www.scottaaronson.com/democritus/