Truth Social I'm sure. As hard as it is to believe, there are still people who think Trump is a genius, and still want to "own the Libs".

Hopefully of course, the AIs will be given the leeway to ignore the scripts (not the actual script items, just the forced structure). I worked in a tech support call center many years back at this point and I always got pretty low marks on the metrics because I didn't follow the scripts and I also had long call times because I just solved their problems and didn't pull every dirty trick in the book to punt them and get them to call back like the agents with good metrics. The scripts, of course, are always stupid. Anyone with a half decent handle on the problem has already covered most of the script with their initial description of the problem, so being forced through steps they have already taken, etc. can be very difficult. You do have to be able to figure out whether or not they actually did some things they said they did of course, but most support is done via remote now anyway, and an AI can check logs and so forth to figure that out pretty easily and quickly.

Honestly, can't say I disagree with this as a solution for first tier at least at this point. As others have pointed out, this can potentially end phone menu trees. Not to mention that they will likely allow the AI, as an expert system, to actually not use ridiculous call center scripts. I still absolutely think that you need human problem solvers to escalate to, but the first people you usually get on the phone are often not problem solvers. I was on the phone for support with a fast charging station the other day, we were past half an hour on the phone call after many, many "please wait a moment sir" and long silences interspersed with requests for basic information that should have been gathered in the first minute (it was at least 20 minutes in before they bothered to establish my location, the name of the charging port I was at, etc.). Basically, by the time they actually got to the point of actually wanting to try something, I had just gotten a text from my family to come and pick them up. My guess is that, aside from having no idea what they were supposed to be doing, they were also probably overloaded like many call center workers. Probably answering not only chats as well as phone calls, but also quite possibly handling multiple phone calls at once as well.

AIs can handle multiple calls at once without making you wait on hold, and they can get to the point pretty quickly and, if they have the right training data, they aren't clueless about what their job is. So, a well done AI is going to be vastly better than most of the humans you encounter at many call centers. Of course, that's no substitute for the experienced human that you can eventually get after dealing with a clueless human for a while, but the AI can still escalate you if it can't help.

Then, there's the communication issues. AI voices these days tend to be more understandable than humans. Not just because they can be made to speak pretty much any language or assume an accent that is clearly understandable to the listener. I don't just mean because of foreign language speakers either, people with heavy regional accents who speak your own language can be just as hard to understand sometimes over the phone. Not to mention, the AI has no need to constantly put you on hold and off (during which humans are usually cutting off parts of their own message), it doesn't have background noise, it isn't using a microphone that might not work well, and just in general, there are less physical, electrical, and digital layers involved in getting their voice to you. Then going the other way, we have gotten to the point where, though AI might not be better at picking out what you are saying than the best human, it's better than the average human.

All that said, AI is definitely not ideal at this point. It's just probably better than the average call center frontline human if it's properly prepared.

Look, I hate to even write this because the apologism (to paraphrase "Yeah, but it wasn't all that violent") should invalidate what you're saying automatically - sure, the degree might matter, but it's still clearly over the line - but the summary did use the term "violently smashed". Now, you could say that's exaggeration, or there's another side of the story, etc. What I can tell you anecdotally, though it's not about a creampie, is about the time someone smashed an egg on my back. Or, that is, they tried to, but they hit pointy end (and these were also farm fresh eggs with much stronger shells than modern factory-farmed eggs) first where the eggs have tremendous structural strength, so it was like being hit by a pointy rock that stayed completely intact and then fell and smashed on the floor. The idea that the thing is just soft and harmless can be a misconception. The same goes for a pie. It depends somewhat on whether the pie tin is foil or solid metal, for example. But also on how hard the person is swinging the pie and whether they keep pushing and grinding it in, etc. after (because you know some people really, really will, and quite hard). Basically, you're talking about a punch in the face, barely cushioned by the cream pie. Whichever way you look at it, this is hazing and bullying and it's wrong.

See, this is one of those "quiet part out loud" things. Where a little outburst pretty much proves that your previous claims that this sort of thing is "good natured" is quite simply a lie. It pretty much is all about singling someone out for abusive bullying for you.

How are these people not in prison?

How about "Take the guns first, go through due process second" from Donald Trump. Also not previous typically a Republican position (except when the Black Panthers emerged, I suppose).

Regardless of exactly how rough or not the estimate is, The math does make it clear that you don't want to put it under your pillow. If it's 0.08 mrem for five minutes, then it would be .96 mrem for an hour and 7.86 mrem for a nights sleep. Then about thirteen nights to hit the yearly max of 100 mrem. Basically, in a year, if it's where you sleep, you'll get double your yearly dose. Wear it as a necklace, even to bed and you're looking at quadrupling your yearly dose. However you want to nitpick about how much radiation it takes to increase your cancer risk by whatever percent, eventually you'll hit it.

Seriously? Basically all life on Earth as we know it (with the exception of stuff like chemosynthetic life) ultimately gets its energy from the sun. When I was a kid, they taught this stuff in kindergarten. All other things being equal (and subject to some other factors, but not ones that greatly invalidate this general rule) crop yields are directly proportional to the amount of light we get from the sun. Dim that enough to prevent global warming, and we can't feed ourselves any more without using a lot more cropland. Not to mention that most farms operate on incredibly thin margins as it is, so there's a mass wave of them going out of business, etc.

The greenhouse effect doesn't happen because we get too much sunlight, it happens because the atmosphere is basically transparent to most of the incoming light but, when it gets converted to IR, the atmosphere is opaque to that IR, so it absorbs it and warms and re-emits it back to the ground. The amount of light passing through the atmosphere is a completely different variable. Trying to fix global warming by tweaking a different variable that affects other things too is a very tricky proposition. It's like if the water pressure in a downstairs tap is too high, but it's fine on the third floor, so you put in a regulator at the water main and reduce the pressure for the whole house. Great, you just solved the downstairs water pressure problem, but you just created a low water pressure problem on those upstairs taps. Basically, you will have "fixed" the wrong problem. If we want an analogy that relates more directly to heat, let's say you have an electronics project that keeps overheating. So you "solve" the problem by reducing the amperage from the power source. With less power it doesn't heat up so mach. Problem solved, except of course that now for some strange reason the LED display isn't lighting up right and it keeps powering on, then abruptly powering off and back on again. Weird.

I didn't even go into weather. Dimming the sun on a global scale will have a massive effect on weather. I mean, that's the idea right? To affect climate, which will also affect weather. The problem is that it won't just magically be the same weather you would get without the greenhouse effect, the actual distribution of energy will be different. Among other problems will be severe droughts in various areas that will increase desertification. Basically, you'll break monsoon cycles in some areas, decrease the hydrologic cycle overall, etc. If you're shaking your head and confused about how some of the same bad stuff from global warming can happen even in a system that reduces overall warming, that's one of the reasons why many have been trying to emphasize that it's about climate change. All of the Earth's biospheres as well as human civilizations tend to rely on the status quo. Massively changing it on the short term is usually a catastrophe. Global warming does that, but so does global dimming. You can't just target one single variable (global average temperature) and if you keep that stable, you're golden.

But we don't want that! Yes, we want the global temperatures to stay down, but not by having less sunlight. That would have its own set of problematic consequences.

And there are 240 pence in a pound, so that's 960 Farthings. Four farthings in a pence, 12 pence in a shilling, 20 shillings in a pound. Of course, then it gets more complicated when you get to coins like the guinea, which was introduced as a one pound gold coin, but since it was gold, was prone to fluctuating values, but the British government eventually locked its value at 21 shillings. Later, in the Regency era, they minted the sovereign, which had less gold to reset the value and that was worth exactly one pound. Of course, both coins were vulnerable to the bullion value being too far above the face value to avoid the coins being melted down. So, while the system isn't impossibly complicated, it wasn't very cut and dried either.

It's only rocket science. It's not like imperial to metric conversion errors are going to crash something into Mars or anything.

That may be the case. That's why they were "wondering". The answer in fact could even be that time will tell if the risk pays off and that it's possible that, even if all three are successes, it still might actually not have been worth it vs. three new satellites on this attempt. Rather, if this succeeds as a proof of concept, it might actually shift from not technically being worth it this time to being worth it on future missions with even more moving parts. In other words, this could actually be an experiment that is worth it just for the knowledge gained.

That is pretty great, and it is one half of what I'm talking about. The other half is on the car side though, where the car would be built so that a self-driving module would be an add on module kit that the car is pre-configured to accept, rather than being essentially monolithically tethered to the car so that it's there when you buy the car (and you actually pay for the hardware) and then you have to buy a subscription. The same should apply to the infotainment, and everything else that ism't really a core feature of the car.

Technically I basically had that some time around 30 years ago. It wasn't the gas pedal though, it was on the steering wheel as part of cruise control.