3000 hours is a lot for a 3D printer, especially if you're not printing in PLA OR you are doing lots of filament swaps OR you're experimenting with different nozzles and filaments.

If you load your printer with one 10kg spool of PLA, odds are good you'll go a while without issues.

Yeah, I think it's been technically possible to automate jetliners for a while now, but when a metal tube with hundreds of people in it develops a technical fault while moving 500+ mph there's no substitute for a pilot.

Probably because there have been lots and lots of news stories about adults leaving keys in the car and children sitting in the driver's seat and accidentally crashing them, since there's no on button.

Oh yes because a kid can press the pedal but not a button.

What is the literal FIRST thing that any child tries to do when you place them in the driver's seat?

No, the Meta board is beholden to shareholders and must work in their interests.

Zuckerberg is kind of rare among tech founders in that he still controls 61% of the company. Therefore the board serves him.

None of this is a secret and this setup is why Facebook is what it is.

> For the safety of any person being accidentally hit

(the spring handles the retraction)

Presumably the motor runs once to extend it and then it locks into position (with some kind of mechanical catch that's calibrated to come loose if you hit anything), rather than being constantly running.

If that's the case, society will inevitably be disappointed.

There are already ten million AI image generators, the overwhelming majority of which do not watermark their outputs. Google auto-inserting them is nice, but ultimately this kind of tool to remove them will inevitably be widespread.

Is there any risk? Don't the model providers also bill by the token?

The accounting could be asynchronous, so you could overshoot your budget by a few requests before you're blocked.

You are a European who drives an EV, so you might not be intimately familiar with just how absurdly large these trucks are. I was recently driving and pulled up at a red light next to one; the top of the hood neatly lined with with the top of the hair of a woman crossing the street directly in front of it. If it had been someone short or a child, the truck driver might have literally been incapable of seeing them.

These things are tolerable on American suburban roads, which are sized generously with even more generous parking. They're an utter disaster for pedestrians because their front profile is more "brick wall" than "gentle slope." It would be fine if they were driven slowly, but the machismo that leads people to buy them also means that they are all absurdly quick.

The methodology is about as sound as a sociology study ever gets - respectable n and in a domain where "only college students" is probably OK.

The conclusion is pretty clear: intelligent people are better at discerning intelligence in others. The effect is weak but statistically significant.

This result lines up with mine (and others) experience, which is that it is significantly harder to bullshit a smart person than a stupid person about your own intelligence. It's fair game to discuss anecdotes in the context of a study like this because human beings don't understand the world in CIs and z-scores, they understand it in narratives.

> That is close to tautological, not some profound discovery about social perception. ;-)

So is my statement above. It would have been a profound discovery if the study found out that intelligent people were no more accurate at judging others' intelligence than anyone else.

The general rule for video codecs is that each release reduces the required bitrate by 30% - a release cycle is 7-10 years so each year the research version of the new codec can be expected to improve ~5%.