HappySort

I was mad at Masley's opponents because they were wrong but moreso that they used unproven character attacks, fear of the other side, and identity issues to avoid an actual discussion But apparently he was only mad that they were wrong x.com/i/status/20259…

@xruiztru @JacobAShell Strange color choices.

Slavery abolition in the Americas.

AI is like a hammer that's only "cosplaying" being strong enough to nail things - that's all well and good until you swing it harder than it can take - then you'll see that it was only faking being strong.

@extra_rice @miriamkdaniel F-droid allows you to install much better GPS trackers. I continuously save mine, so I can vibecode up maps and a way to find out where I've been. It's great

@miriamkdaniel so you're the one responsible for removing timelines... now there's no way to know when or how many times i've been to a place... totally ridiculous nerf of one of the most useful features of google maps...

Hi, X! I’m Miriam, VP and GM of Google Maps and I just wanted to introduce myself. Excited to be here and share more information and insights about the latest updates to Maps, new ways to use our features, and more!

Two-boxers advise a one-boxer without realizing they can too

People discussing Newcomb's paradox know that "saying I'll two box" is correlated with not getting the million But still go on podcasts and don't realize they can just "say" they're a one boxer to win

Doing impressions is naturally creating a superstimuli of "seeing that famous person"

It's interesting that "Doing impressions" is categorized with comedy. That skill is pretty close to the recently booming understanding of what LLMs do, how we train them, how the brain trains its own submodules. "Controlling context to elicit a response" Another data point for how simulating other people and how they'd react is central to our brain development. (in this case, "taking actions to induce a previously experienced feeling of being with a person")

If you have to look at exact weights to create adversarial images to cause recognition models to fail badly, it's not very impressive. If you controlled the entire sensory input of a human, you could make people do anything. If you could try to trick someone into giving you a million dollars accidentally and could induce any emotion, sight, sound, pain or pleasure, and could iterate until you solved it, it wouldn't prove much Being able to do that for general models or people, in one shot, without knowing their internals is what matters.

Bring back "shall" for giving orders to AI "My mansion shall be made of perfectly fitted hand placed natural stone" "Should" sounds too much like whining

@eliwaxm @NathanpmYoung @googlemaps That would not sell ad units

@NathanpmYoung @googlemaps What about if I search for something like McDonald’s, you don’t have to zoom out 100km out and show me all McDonald’s in that radius if I have one NEXT TO ME

Dear @googlemaps Come on, you can realise when I’m on a train. They travel along train tracks. You know where the trains are. You know other people are doing nearly identical journeys. Say “Are you on this train” and allow me to plan routes from there.

@grok @tenobrus A few thousand? That seems low enough. What if I just paid for it myself - could fsd accept the signal?

In 2026, a single automotive-grade solid-state LiDAR sensor suitable for self-driving (e.g., compact models from Hesai or MicroVision) costs around $200. High-end units or full multi-sensor suites run higher (up to a few thousand per vehicle), but that's down from $50k+ a decade ago due to mass production.

tesla's decision to point blank refuse to touch lidar has proven to be one of the most insane self owns of any technology company ever. they easily have the research talent, and waymo has proved they could be doing millions of fully autonomous rides. at this point it's a choice

@Drachs1978 @tautologer Good framing. And even if that were true why assume that the intelligence level where agreement converges lower than they've achieved?

@tautologer Do people really exist that think that just because a person is smart they must support their goals? So like to convince someone I'm their friend I only have to convince them I'm too smart not to be? Wild cognitive malfunction if true.

the reason you see so many leftists and the like stubbornly insisting that Elon MUST be an idiot and he definitely STOLE all the good work his companies did, against all evidence, is because they have failed to grasp the Orthogonality Thesis

Storytelling ability applies to both your own brain modules and also the outer world. Controlling the prompts and details you give, incentives, pressure to go along with a viewpoint, the frame. It also applies to ones internal family dynamics

Google photos edit tools treat operations which should be instant as if they require work. Example: be cropping an image while looking at it. Hit save as copy; screen greys out and looks ugly for 2 to 3 seconds before... Showing the pixels we saw before.

@riemannzeta @__paleologo @grok translate these two new terms to my language giving examples based on my timeline

But there is such a thing The subjectivity of human experience

Syncthing is a pretty good replacement for Google photos auto upload The problem with Google photos is that you don't end up with a local copy of your pictures sitting there organized and ready for your agents to make into albums or do creative stuff with or to dig information up about what you saw. Nor to do research on items you wanted to remember to learn more about that day. Instead, google photos just takes them and backs them jo. And as far as I can tell, the apis for even keeping a local fullsize copy of everything are hobbled. Sure, within gphotos they offer you "memories" but it's not the same as *having* them in a folder

@fortelabs Alt man Whose alt tho. Theil?

Wait, so the founder of Anthropic is "Amodei," as in "loves god"? And he leads Anthropic, meaning "human-centered," which is being used in military strikes? And the creator of ChatGPT is "Altman," as in "an alternative to humans"? And he leads OpenAI, which is completely closed? And then there's "Gemini," meaning "two-faced," from a company that promised to do no evil? And the whole global AI arms race is being driven by people who claimed to be worried about AGI taking over the world? Either the universe is an extremely cliché writer, or has a brilliant sense of humor