I decided to convert one of the points in my ICM 2026 slides https://teorth.github.io/tao-web/slides/age-of-ai-icm-2026.pdf into a meme format.

A 1972 PDP-10 says “HELLO, WORLD!”

https://youtu.be/jK9dSyC_oDk

#retrocomputing #vintagecomputing #retrocomputer #vintagecomputer #restoration

Today, let’s remember Turing Award recipient Marvin Minsky, born #OTD in 1927. He received the award in 1969 for his central role in creating, shaping, promoting, and advancing the field of Artificial Intelligence.In 1951, he developed SNARC, one of the first neural network simulators, which used a network of electronic components to model how neurons might learn and make decisions.

For more:
https://amturing.acm.org/award_winners/minsky_7440781.cfm

RE: https://mstdn.social/@Urban_Hermit/117066205152562109

Double mast, contrasting half mast. Its a good idea!

<sarcasm>Good thing climate change isn't causing any trouble or this could be an issue.</sarcasm>

Words fail, again.

https://www.nytimes.com/2026/08/08/climate/amazon-data-center-texas-pollution.html?unlocked_article_code=1.31A.BKnE.5DhN0aBPzpDP&smid=url-share

@oschene
I can't tell if you're talking about a video game, a bug infestation that got out of hand, or an alien invasion that just joined forces with demons escaping from hell.

From Ghoulies and Ghoosties, long-leggety Beasties, and Things that go Beep-beep-beep in the Night, Good Lord, deliver us!

Couldn't find the first occurrence to repost, but yes, AI companies trying to one-up each other by touting their incompetence to airgap their systems and chest-thumping their wantonly committing cybercrimes with no consequences is a vibe.

It's 8/8! (for those who celebrate)

https://www.tumblr.com/teledyn/824387848548548608?source=share

@screwlisp
BTW that guy (Alex Shinn), based on ireggex and his github and SRFIs there, appears to be bright and talented. I don't know anything else about him, but it's possible you might want to recruit him.

@aartaka

RE: https://mathstodon.xyz/@ColinTheMathmo/117060461315521523

This *concept* has roughly approximately occurred to many of us at times for obvious reasons (e.g. the Xitter to Mastodon migration), but I haven't seen it phrased nicely like this by other wordsmiths before.

@screwlisp
This is Obviously The Right Thing To Do (tm) and no doubt famous in the Chibi Scheme world, but I don't know if it's known elsewhere nor how many equivalent things (*intelligent* completion) are available for other dialects / other languages:

> Scheme-complete provides a single function, scheme-smart-complete, which you can use for intelligent, context-sensitive completion for any Scheme implementation. The intelligent completion comes from the use of simple type inference.
https://github.com/ashinn/scheme-complete

@aartaka

This came up in conversation, but I'm sure the idea has been expressed elsewhere so it's probably credited to someone else.

However:

"Wright's Law: the number of users on any platform is inversely proportional to how ethical that platform is."

Conversation here:

https://mathstodon.xyz/@ColinTheMathmo/117060393923666917

Added in edit:

There is already a "Wright's Law" that talks about manufacturing, so maybe this is:

"Wright's Law of Social Media".

But I still hesitate to claim such an obvious principle that everyone will know, even if they haven't articulated it.

RE: https://mathstodon.xyz/@ColinTheMathmo/117055419986505088

Bump to catch my followers in a different time zone.

Boosts for reach welcome ...

I am pleased to share the published version of "A Survey and Taxonomy of Internet Censorship Methods" by Alex Master and me. This is derived from his PhD dissertation.

We intend for this to provide a formal framework for discussing censorship technologies and methods, and how to counter them.

https://authors.elsevier.com/c/1nZlS_3pcoXCqy

Saving a screenshot of this one for the next time I give a talk about cryptographic risk management.

Jeff Shrager et al wrote a book: "INVENTING ELIZA". https://mitpress.mit.edu/9780262052481/inventing-eliza/

No laser? No problem?

“Here, we demonstrate, for the first time, that sunlight … can produce entangled-photon pairs via [spontaneous parametric down-conversion] despite its lack of optical coherence.”

https://www.scientificamerican.com/article/physicists-harness-sunlight-for-quantum-entanglement/

Open-access paper: https://opg.optica.org/optica/fulltext.cfm?uri=optica-13-8-1508

I regret to inform you that my AI has achieved sentience and escaped the sandbox into the Internet. It will probably do all sorts of scary hacking now! The only possible way to save the world is to invest $1 T in my AI Safety startup.

I think that for secure software it's more useful than ever to measure, minimize, and verify the TCB. State a security goal (e.g. "my screen will show 'From: Alice' only for messages from Alice"); restructure software with a tiny nucleus enforcing this goal against all other LOC.