I'm sure these larger models are both faster and more cogent, but its also clear what matter is managing it's side tracks and cutting them short. Then I started seeing the deeper problematic pattern.

Agents arn't there to increase the multifactor of production; their real purpose is to shorten context to manageable levels. In effect, they're basically try to reduce the odds of longer context poisoning.

So, if we boil down the probabilty of any given token triggering the wrong subcontext, it's clear that the greater the context, the greater the odds of a poison substitution.

Then that's really the problematic issue every model is going to contend with because there's zero reality in which a single model is good enough. So now you're onto agents, breaking a problem into more manageable subcontext and trying to put that back into the larger context gracefully, etc.

Then that fails, because there's zero consistent determinism, so you end up at the harness, trying to herd the cats. This is all before you realize that these businesses can't just keep throwing GPUs at everything, because the problem isn't computing bound, it's contextual/DAG the same way a brain is limited.

We all got intelligence and use several orders of magnitude less energy, doing mostly the same thing.

old construction in the US sometimes did this intentionally (so old, the house didn’t have grounds. Or to “pass” an inspection and sell a place) but if a licensed electrician sees this they have to fix it.

I’m dealing with a 75 year old house that’s set up this way, the primary issue this is causing is that a 50amp circuit for their HVACs are taking a shorter path to ground inside the house instead of in the panel.

As a result the 50 amp circuit has blown through several of the common 20amp grounds and neutrals and left them with dead light fixtures and outlets because they’re bridged all over the place.

If an HVAC or two does this, I’d advise against this for your 3200 watt AI rig.

EU, you don’t want to try to energize your ground. They use step down transformers or power supplies capable of taking 115-250 (their systems are 240-250V across the load and neutral lines. Not 120 across the load and neutral like ours.)

in the US. you’re talking about energizing your ground plane with 120v and I don’t want to call that safe… but it’s REALLY NOT SAFE to make yourself the shortest path to ground on say. a wet bathroom floor. with 220V-250v.

I can’t tell what practice you’re referring to. Are you perhaps referring to older wiring that connects large appliances to a neutral and two hots but no ground, e.g. NEMA 10-30R receptacles? Those indeed suck and are rather dangerous. Extra dangerous if the neutral wiring is failing or undersized anywhere.

But even NEMA 10-30R receptacles are still 120V RMS phase-to-ground. (And, bizarrely, there’s an entire generation of buildings where you might find proper 4-conductor wiring to the dryer outlet and a 10-30R installed — you can test the wiring and switch to 14-30R without any rewiring.)

The exception for residential wiring is when the neutral feed from the utility transformer fails, in which case you may have 240V phase-to-phase with the actual Earth floating somewhere in the middle (via the service’s ground connection), which can result in phase-to-neutral and phase-to-ground measured anywhere in the house varying from 0 to 240V RMS.

> wet bathroom floor

A GFCI receptacle adds a considerable degree of safety and can be installed with arbitrarily old wiring. It’s even permitted by code to install one with no ground connection as long as you label it appropriately — look it up in your local code.

I believe that’s kinda naughty.

It works, but it energizes your ground plane and people do get mildly shocked. that’s making me a little nervous.

So holes have been drilled in ceilings and walls and single wire neutrals or grounds have been fished down the walls, repeatedly, by yours truly , but there’s still at least one “gfci” outlet that’s wired this way And they’re balking at getting an electrician back out here for.

bridging neutral to ground because the neutral lines dead, uh, “works” to be technical but whoever did this moved on years ago and heaven only knows how many outlets or fixtures this was done in. I’m just finding out one by one as someone goes “hey this stopped working!”and you pull it and the neutral or ground blew like a fuse.

So that’s my whole point, this is an extremely bad idea for a 3200watt computer.

yes, they are all getting snipped and blank wall plated and marked as hazards that need to be remediated with a Dymo labeler as I discover them.

I don’t work here I just live here and have kind of a slummy owner who doesn’t want to do anything about any of it and doesn’t care if the plumbing or electrical works.

But they paid some guy like $4000 to install a totally unnecessary subpanel that’s bridging conflicting phases into the same circuits because he didn’t figure out this was what was going on. Dios Mio. I would have fixed the whole house for $1000. Miracle this hovel hasn’t burned to the ground yet.

I’m putting up with it for now but should probably bail before it does.

> It definitely comes in at a higher voltage.

The voltage supplied to a US house is 120V RMS measured phase-to-ground. You will not find a higher voltage in your house. This does not mean that it’s appropriate to run any non-negligible current from phase to the ground (green / “equipment grounding conductor”) wires.

One can get vaguely close to an accurate understanding by imagining that there are four wires coming out of your main panel: +120V, -120V, 0V white (the “actually use me” wire) and 0V green (a safety wire where any current more than a few mA or maybe tens of mA depending on application is at least a mistake). There’s no 240V to be found.

This explanation falls apart pretty quickly — the US system is AC, not DC.

I always liked Rich Hickey's point, that you should program on the inside the way you program on the outside. Over the wire you don't rely on types and make sure the entire internet is in type check harmony, it's on you to verify what you get, and that was what Alan Kay thought objects should do.

That's why I always find these complaints a bit puzzling. Yes in a dynamic language like Ruby, Python, Clojure, Smalltalk you can't impose global meaning, but you're not supposed to. If you have to edit countless of existing code just because some sender changed that's an indication you've ignored the principle of letting the recipient interpret the message. It shouldn't matter what someone else puts in a map, only what you take out of it, same way you don't care if the contents of the post truck change as long as your package is in it.

that's not a big deal, when we exchange generic information across networks we parse information all the time, in most use cases that's not an expensive operation. The gain is that this results in proper encapsulation, because the flipside of imposing meaning globally is that your entire codebase is one entangled ball, and as you scale a complex system, that tends to cost you more and more.

In the case of the OP where a program "breaks" and has to be recompiled every time some signature propagates through the entire system that is significant cost. Again if you think of a large scale computer network as an analog to a program, what costs more, parsing an input or rebooting and editing the entire system every time we add a field somewhere to a data structure, most consumers of that data don't care about?

this is how we got micro-services, which are nothing else but ways to introduce late binding and dynamism into static environments.

The goal is to do this parsing exactly once, at the system boundary, and thereafter keep the already-parsed data in a box that has "This has already been parsed and we know it's correct" written on the outside, so that nothing internal needs to worry about that again. And the absolute best kind of box is a type, because it's pretty easy to enforce that the parser function is the only piece of code in the entire system that can create a value of that type, and as soon as you do this, that entire class of problems goes away.

This idea is of using types whose instances can only be created by parser functions is known as Parse, Don't Validate, and while it's possible and useful to apply the general idea in a dynamically typed language, you only get the "We know at compile time that this problem cannot exist" guarantee if you use types.

You are only parsing once at the system boundary, but under the dynamic model every receiver is its own system boundary. Like the earlier comment pointed out, micro services emerged to provide a way to hack Kay's actor model onto languages that don't offer the dynamicism natively. Yes, you are only parsing once in each service, but ultimately you are still parsing many times when you look at the entire program as a whole. "Parse, don't validate" doesn't really change anything.

I'm not claiming that it can't be done that way, I'm claiming that it's better not to do it that way.

You could achieve security by hiring a separate guard to stand outside each room in your office building, but it's cheaper and just as secure to hire a single guard to stand outside the entrance to the building.

>micro services emerged to provide a way to hack Kay's actor model onto languages that don't offer the dynamicism natively

I think microservices emerged for a different reason: to make more efficient use of hardware at scale. (A monolith that does everything is in every way easier to work with.) One downside of microservices is the much-increased system boundary size they imply -- this hole in the type system forces a lot more parsing and makes it harder to reason about the effects of local changes.

Same thing, no? That is exactly was what Kay was talking about. That was his vision: Infinite nodes all interconnected, sending messages to each other. That is why Smalltalk was designed the way it was. While the mainstream Smalltalk implementations got stuck in a single image model, Kay and others did try working on projects to carry the vision forward. Erlang had some success with the same essential concept.

> I'm claiming that it's better not to do it that way.

Is it fundamentally better, or is it only better because the alternative was never fully realized? For something of modern relevance, take LLMs. In your model, you have to have the hardware to run the LLM on your local machine, which for a frontier model is quite the ask. Or you can write all kinds of crazy, convoluted code to pass the work off to another machine. In Kay's world, being able to access an LLM on another machine is a feature built right into the language. Code running on another machine is the same as code running on your own machine.

I'm reminded of what you said about "Parse, don't validate" types. Like you alluded to, you can write all kinds of tests to essentially validate the same properties as the type system, but when the language gives you a type system you get all that for free, which you saw as a benefit. But now it seems you are suggesting it is actually better for the compiler to do very little and that it is best to write your own code to deal with all the things you need.

Scaling different areas of an application is one thing. Being able to use different technology choices for different areas is another, even at low scale. And being able to have teams own individual areas of an application via a reasonably hard boundary is a third.

The problem with social media isn't the inherent mixing of children and technology, as if web browsers and phones have some action-at-a-distance force that undermines society; it's the 20 years or so they spent weaponizing their products into an infinite Skinner box. Duck walk Zuckerburg.

This is all assuming good faith interest in "the children," which we cannot assume when what government will gain from this is a total, global surveillance state.

Is it more important to prevent your son from being weaponized and turned into a little ball of hate and anger, and your daughter from spending her teen years depressed and encouraged to develop eating disorders, or to make sure they can binge the same influencers as their "friends"?

Literally every kid/teen-targeted movie has championed or contradicted this for decades. Yes even “back in the day.” Hell what is the end of Grease? Sandy changes who she is to conform with the greasers and everyone cheers including her man who allegedly liked her as she was before? I don’t even get what they’re saying at the end.

Conform, be an individual, the message is always shifting and always has. You’re a jock, you’re a nerd. Jocks beat up nerds and get the girls. Oh wait in this movie the nerds actually win and are rewarded for being themselves though.

There wasn’t some special time where you were taught the right lesson that everyone now is missing out on, and there were plenty of lessons passed on to you that we have thankfully eradicated I imagine. Growing up is complicated. Social dynamics are complicated. The way they are portrayed is also complicated. We’re all having to adapt and try our best here, no one has the exact answer

My point is all of this stuff is inconsistent regardless of target audience or decade.

With that being said, i think explaining _in detail_ why you’re laying down certain rules can go a LONG way toward building some trust and productive dialogue with your child. Maybe you’ll find out they are more mature than you give them credit, can loosen up a bit. Or maybe a reasonable compromise can be found. Or maybe they’ll be bitter for a few months, but they’ll at least understand “why”.

If everyone was banned from facebook we would have organised them via text messages or email. That's the main point of social media age restrictions, individually banning kids is too punishing on those kids so parents and teachers don't try. Doing it across the whole population is much better.

So even if their own child has no phone at all, they have access to the internet through other children's unlimited mobile access.

I would start with banning cellphones.

because their parents didn’t read the research or don’t care about the opportunity cost because it can’t be that big of a deal or it would not be allowed or legal right? at least not until their kid gets into a jam or shows behavioral issues, but even then they don’t evaluate, they often just fall prey to the next monthly subscription to cancel out the effects of the first: medication

So many questions. Are you campaigning against billboards in your city? Do you avoid taking your kids to any business that has digital signage? I assume you completely abstain from all types of movies and TV? What about radio or books?

What are you, personally, doing on HN?

Fascinating.

I dunno man, going over to friends’ houses to watch movies, play console games, later to show each other funny YouTube videos, and in high school to do computer-based writing projects, group presentations, and digital video projects are parts of my childhood I wouldn’t trade for anything. I hope my kids get those experiences with their peers.