I'm growing to hate social media. This last week I've been using a browser plugin to just block facebook, twitter and instagram due to the fact its just a sea of people hating and accusing each other of psychotic bullshit over the CK murder and like...... get a fucking grip americans.... Its just bad for the soul to be flooded with hyper partison aggravated political bullshit all day. So I've blocked it, and despite my inner idiot wanting to get on there and yell at people, I'll take the dopamine loss and just do something else. Like play the piano, or run the storyline on that new Dune Awakening expansion. Or go and drink rum with friends. Or fucking anything other than doomscroll on facebook again.

If AI kills social media , then maybe AI might just be good for something.

Some experiments show that light doesn't exist unless we look at it.

No experiments show that. Its just a strange interpretation of certain features of photons. But to be clear that doesnt mean "the photon is not real", rather it shows probabilistic properties under specific circumstances. Strange does not mean fake.

as for whether it 'experiences' time. Well thats a complex question. Due to time dilation , from *its* inertial frame, one would things happen pretty much instantly, it is effectively always at the end of its own time cone. But from an observers inertial frame, it would be travelling at precisely the speed of light (in a vacuum).

These are, I should note, entirely different things, and it would muddy the mind to mix these two ideas together.

One day we will learn in school that empathy is bad. An out of date emotion that was only useful when we were a grazing herd.

Spoken like a true sociopath.

Without empathy we dont have society. Without empathy we don't have morality. Without empathy were little more than apes dashing each others brains out with rocks over territory.

What a stupid claim. Evolution gave us the mirror neuron for a goddamn reason.

"Python is one of the worst offenders when it comes to breaking changes"

Eh? the 2 -> 3 transition was well over a decade ago dude, and the API stability of python libraries has always been one of its selling points. Considering its rivals like JS and its never ending library updatge hell, thats a weird claim to make

Why have one authoritarian country have all the AI when 3 authoritarian countries can have all the AI.

(Sorry americans, but right now I'm counting you guys as an authoritarian country. With that said, if you guys arent using that constitution anymore, can we have it? )

One of my more bonkers moves as a 20something in the 1990s was busting a girlfriend out of a mental ward she was being held involventarily in. she was actually fine , but some asshole psych had her basically captive because apparently not complying with her turbo christian parents is madness.

So after an afternoon at the pub, me and a bunch of the boys just barged in and walked her out the front door with the "We are bigger and scarier and you have no power over us" glare.

The wild thing is ..... it worked. The cops didnt come for her, and to this day 30 years later she's never heard a thing from the ward. I've long suspected there was something very wrong going on in that hospital and they decided it better to just let her go than have the cops involved.

Thats usually where these sorts of accidents happen in my experience. I ride an escooter, and here in australia the speed limits 25km/h on an escooter. Thats a pretty good limit. At 25km/h most crashes are just ouchies. Scratches, bruises, but rarely anything more serious than that unless you collide with a car or human. If the cops catch you with a scooter going faster, they can confiscate it and under anti hooligan laws (originally designed to go after illegal drag racers) can confiscate and destroy the scooter. I think thats entirely reasonable. Break the law, get the penalty. cause -> effect.

The problem is all the scooter stops advertise these ridiculously souped up scooters that let you switch off the govenors and can get upwards of 70km/h. Even 35kmh your starting into broken bones territory. Get thrown face first over the handlebars at 70km/h, you'll be lucky if you come out of that a tetraplegic.

These vehicles are great. I spend less per year on electricity to run it than most electric cars will spend in a weekend. Its as close to carbon zero as you'll get without going full fred-flinstone and just relying on feet power. But these dickhead shops selling people 70km/h scooters with the govenors disabled and not teaching people even the basics in how to ride safely and road rules, its gonna get the whole damn lot banned.

Correct. A higher mass means a bigger event horizon.

The formula for this is the Schwarzschild radius, r = 2GM/C^2 where G is the gravitational constant, M is mass and C is the speed of light. So yes, the higher the mass, the bigger the event horizon.

Which maps neatly onto this observation that when the blackholes merged the resultant blackhole was not smaller than the radius predicted by adding the two masses together.

Though its more a vindication of Schwarzschild since the conclusion flows naturally from this fairly basic law.

They got more self-driving cabs than Tesla, and Tesla is getting hundreds of billions in valuation from their "Robotaxi".

In either case, if they can scale this there's loads of money waiting. I think they're currently Waymo's only real competitor.

It's been more than a meme.

As the article points out, it's been a question in philosophy since antiquity. At least 2500 years of discussion, probably more. We have no way to know how different people process what they see without peeking inside the brain and comparing. This is newsworthy as it's the closest we've come to verifying it.

There are still open questions about perception and interpretation in addition to just neural pathways, particularly around those with different sensitivities, but that's at least a start.

Color blindness missing one, two, or three sensitivities, tetrachromats or having a fourth sensitivity, shifted sensitivities that peak at slightly different places for different people, all of them lead to ways the colors could be perceived and interpreted differently by different people. It's a good start at research, but there's a lot more that can be answered.

The thing is, it sort of was in its early days. When facebook came out, it really was just your friends, and the events they where throwing. Facebook was still in the "We have no idea how to monetize this, so we just scale for now" phase of pre-enshitification. And prior to that, we had myspace and a few things that where... interesting but not quite the full social media yet, and prior to THAT we had the original social media, IRC and Usenet, and yeah that was never really enshitified as it didnt belong to anybody. and somewhere in there we had bulletin boards, of which /. was a more news oriented example of.

Personally I think the entire social media thing needs to be burnt to the ground and the sand salted so nothing grows in that soil again. Its doing too much damage. But I dont know how we'd achieve that without authoritarian nonsense like bans, and well we can have a look at nepal to see how well that goes.

If you spend time with the higher-tier (paid) reasoning models, you’ll see they already operate in ways that are effectively deductive (i.e., behaviorally indistinguishable) within the bounds of where they operate well. So not novel theorem proving. But give them scheduling constraints, warranty/return policies, travel planning, or system troubleshooting, and they’ll parse the conditions, decompose the problem, and run through intermediate steps until they land on the right conclusion. That’s not "just chained prediction". It’s structured reasoning that, in practice, outperforms what a lot of humans can do effectively.

When the domain is checkable (e.g., dates, constraints, algebraic rewrites, SAT-style logic), the outputs are effectively indistinguishable from human deduction. Outside those domains, yes it drifts into probabilistic inference or “reading between the lines.” But to dismiss it all as “not deduction at all” ignores how far beyond surface-level token prediction the good models already are. If you want to dismiss all that by saying “but it’s just prediction,” you’re basically saying deduction doesn’t count unless it’s done by a human. That’s just redefining words to try and win an Internet argument.

You can't say he didn't boldly go.

They do quite a bit more than that. There's a good bit of reasoning that comes into play and newer models (really beginning with o3 on the ChatGPT side) can do multi-step reasoning where it'll first determine what the user is actually seeking, then determine what it needs to provide that, then begin the process of response generation based on all of that.

This is not a surprise, just one more data point that LLMs fundamentally suck and cannot be trusted.

Huh? LLMs are not perfect and are not expert-level in every single thing ever. But that doesn't mean they suck. Nothing does everything. A great LLM can fail to produce a perfect original proof but still be excellent at helping people adjust the tone of their writing or understanding interactions with others or developing communication skills, developing coping skills, or learning new subjects quickly. I've used ChatGPT for everything from landscaping to plumbing successfully. Right now it's helping to guide my diet, tracking macros and suggesting strategies and recipes to remain on target.

LLMs are a tool with use cases where they work well and use cases where they don't. They actually have a very wide set of use cases. A hammer doesn't suck just because I can't use it to cut my grass. That's not a use case where it excels. But a hammer is a perfect tool for hammering nails into wood and it's pretty decent at putting holes in drywall. Let's not throw out LLMs just because they don't do everything everywhere perfectly at all times. They're a brand new novel tool that's suddenly been put into millions of peoples' hands. And it's been massively improved over the past few years to expand its usefulness. But it's still just a tool.