I bet if such a law existed in less than a month all those AI developers would very quickly abandon the “oh no you see it’s impossible to completely avoid hallucinations for you see the math is just too complex tee hee” and would actually fix this.

Nah, this problem is actually too hard to solve with LLMs. They don’t have any structure or understanding of what they’re saying so there’s no way to write better guardrails… Unless you build some other system that tries to make sense of what the LLM says, but that approaches the difficulty of just building an intelligent agent in the first place.

So no, if this law came into effect, people would just stop using AI. It’s too cavalier. And imo, they probably should stop for cases like this unless it has direct human oversight of everything coming out of it. Which also, probably just wouldn’t happen.

My experience is dated, but figured I’d share it in case no one else has any input.

I owned a few Motorola Android phones before and after the Google involvement. I think my most recent purchase was 2015.

At that time, they were extremely “pure android” with very few additions beyond the stock experience. The things they added were way ahead of their time - I still think those devices had the best “always on” display implementation to this day, and they did it way before it became a norm.

Their software and update support was rivaling Google at the time, and most other manufacturers were still in the days of 2 years of updates if you’re lucky.

They just stopped making phones it seemed like. I ended up moving towards Pixels over the years, but Moto is the one company that would tempt me to switch back. That or maybe HTC but they’re dead.

Hope you get a more recent answer - I didn’t even realize they were still making phones to be honest.

Yes, because when you run systems like that, you use the AI, and you have the people as a fallback for when the AI fails.

It was primarily watched by people in India because the AI was failing the vast majority of the time.

So yeah, the state of the art AI is… Failing at its job 70% of the time. Instead of the hoped goal of 5%.

That’s more “device” pairing than “parts” pairing. The thermostat to HVAC communication is a standard. Sure, if someone started forcing that, that’d be bad. But that’s more akin to Apple’s “iOS only works with MacBooks” type shit with Airdrop and such than it is to their “you can’t replace the camera in your phone unless it’s from us”. They’re both problems, but the one you’re describing is both not happening and a different issue. I’m not saying it won’t happen but it’s a different topic.

What… The… Fuck?

If your thermostat could cause a fire or gas leak, your HVAC system is flawed. This is entirely a fabricated concern. If anything, I’d chalk it up as reasons why maybe right to repair the HVAC isn’t a great idea. A properly setup HVAC wont let anything tell it to do that.

Firstly, I said this one was iffy to me.

Second, the subtopic was HVAC and thermostats are like, the electronics that control the HVAC which I wouldn’t even really necessarily bucket into HVAC. It’s like HVAC adjacent.

Third, this whole topic is about right to repair, not right to replace. So the on topic argument is “you want to be able to repair the same thermostat with off brand parts”, to which I say, yes? Probably? I don’t see how that’s a problem.

And fourth, who the fuck would buy an Amazon thermostat, lmao.

Yeah, I just meant for explaining the function of what the thing does.

I mean, I don’t want the thing supplying the air I’m breathing to accidentally not burn all the gas and lead to carbon monoxide poisoning etc… Things like the ductwork and shit, for sure, but not like, a burner.

It’s basically an analog version of an HDMI cable. Except no audio, only video.

It’s like the yellow RCA cable, but for computer monitors instead of TVs

I could see an argument about medical devices, HVAC, and vehicles… But I don’t think I’d agree with them. Except maybe medical.

Consoles and toothbrushes though? What the fuck?

(Plus the Zojirushi has no timer, no status indicator, and no power button. To turn it on you plug it in.)

Every zojirushi I’ve ever seen has more buttons and settings than most microwaves. Did you buy the cheapest one they sold?

There’s isn’t enough physical space for three sensors on a smaller phone especially if it’s the size of the iPhone mini

I wouldn’t go as far as to claim that “more cameras” is the complaints being made here. Sure, telephotos make sense as things that take up more space. But most people are using them for like 1 in 50 shots or something. I have an extremely hard time believing that someone would genuinely notice the difference unless they’re an extreme case or they’ve been told the other ones are better. Within reasonable effective focal lengths, these are pretty negligible in the sizes we’re talking about.

If Apple couldn’t make a smaller phone sell particularly well, I doubt anyone else could.

I hard disagree with this. Apple is literally the worst company to try to make this shit work. Apple’s core selling point is the status symbol of it all. People trying to show off having the flashiest phone are not going to buy a product being touted as a half baked smaller and cheaper version of something else. Their entire marketing was about it being mini. Apple customers are not the core audience for something like this, and Apple marketed it as exactly what people disliked about small phones.

around or less than 5% of total iPhone 12 and 13 sales

I find it more surprising that this was below expectations than I do that only 5% of people bought a smaller phone. I doubt much more than 1 in 20 people really is after a smaller phone. I’m sure they exist, but based on the people I know and the number of people I’ve heard interested in smaller phones, I’d estimate it more like 1 in 20 to 1 in 40. It’s not for most people by any means. But 1 in 20 is still a decent number of people.

They have to understand that the cameras on the biggest flagships occupy a lot of space and it isn’t feasible to bring it to a smaller form factor.

Not… Really… Sure it makes some difference, but the much more constraining factor is the money. Cameras arent that big, but they’re one of the priciest pieces of hardware in the device.

The problem is more that they keep trying to sell small phones at cheaper price points. So they end up with much worse screens, socs, and cameras so they perform like shit. People don’t want a small phone because they don’t care about their phone. People want small phones because the standard size is fucking huge. They need to make a high-ish tier small phone instead of low tier small phone that performs like the 50 Walmart shit.

I know it’s not everyones cup of tea but…

I ended up buying a Pixel Fold for the Fold aspect. There’s an added benefit that it’s really short when folded so I can reach the entire screen one handed. It’s heavier and thicker than a slate phone, but I actually enjoy the folded experience way better than other phones.

You’re entirely missing the point.

The requirements and basis of IQ tests are they are problems you haven’t seen before. An LLM works by recognizing existing data and returning what came next in the training set.

LLMs work directly in opposition of how an IQ text works.

Things like past experience are all the shit IQ tests need to avoid in order to be accurate. And they’re exactly what LLMs work off of.

By definition, LLMs have no IQ.

That’s because LLMs aren’t intelligent. They’re just parrots that repeat what they’ve heard before. This stuff being sold as an “AI” with any “intelligence” is extremely misleading and causing people to think it’s going to be able to do things it can’t.

Case in point, you were using it and trusting it until it became very obvious it was wrong. How many people never get to that point? How much has it done wrong before then? Etc.

Yes, but Chromebooks are far from “newcomers” these days. They’ve been out a while. Many people who grew up using them in schools are now making their own purchasing decisions, etc.

Like he said as the second sentence of his comment…

You’ve never worked in enterprise then.

These solutions are skipping the majority of the core problems he mentioned. And even the problem you’re trying to solve here isn’t even fully solved by this solution. You’re taking a narrow sliver of one point in his argument and arguing about that and just tossing out the rest. Even if we accepted your proposal, Linux still isn’t enough of an answer here.

While I don’t really disagree, look at the market share of Chromebooks. If “most people” only needed internet access, “most people” would be on Chromebooks by now. It’s not like they’re unknown anymore.

Yeah, and it’s likely way less costly to the company to just buy a new win 11 computer than it is to pay an employee to train on new software. Not to mention the cost of paying someone to find someone to do a Linux conversion, paying the person doing the conversion, and the loss of productivity as the person learns. Not to mention the cost of changing IT infrastructure, hiring new IT people to manage those machines, etc.

There’s a reason companies don’t just switch at the drop of the hat. There’s too much commitment and institutional knowledge already and moving is not a simple change.