aetherson

I'm a parent and while I invite people over (indeed, this last Sunday we had another couple and their kid over) and do this thing, very few of our parent friends appear to do it. Or I guess maybe they do it with each other and just don't invite us, but I think it's rarer than you think.

She's probably not? If she's gone two and a half years since her gallbladder was removed, and years before that when sge had gallstone symptoms but thought they were something else, that was before widespread use of GLPs.

I believe they wouldn't have to because of the stepped-up-basis, but my point is that even if somehow they were paying the full 40%, it's still an inconceivably large amount of money.

Even very large salaries are probably not a very significant amount of wealth for these people. I think this is a case where "all large numbers look the same."

If Warren just directly gave his money to his three children equally, and paid 40% inheritance tax on that money, and then the children each just used 1% of that wealth per year, then they'd each get:

144000000000 * 0.6 / 3 / 100 = $288 million each year.

No foundation pays its CEO 10% of that much money. I think we may have to accept that in fact this is a way to give money to charitable causes, not just horde it.

What else would his repute be based on? Maybe telling retrospective truth? Per the article, he does terribly by that metric as well.

You could imagine that pre-diagnosis confusion relating to Alzheimer's increases the risk of traffic accident and kills taxi drivers, too. Might not operate for truckers as much because they're more likely to survive accidents.

18.

Okay, if it's bad for the price they're offering, use the more expensive API. If they themselves can't afford it I don't know what you want here.

That was very much on the table, though not locked in.

The US's population is not declining. However, without immigration, our population growth would be very close to zero, and a yet lower rate of natural increase is locked in over the next couple of decades (more old people who will die than children).

And, I mean, it's obviously hard to predict beyond that, but it doesn't seem like anyone has any real clear answer to the trend of steadily decreasing TFR right now.

I mean, he was pretty vindicated on Uber. Which kind of hurts me to say, I was a long time Uber bear. But it did indeed emerge from the pandemic stronger.

Sure. But that's not the subject.

They learn during training, which is what we're talking about.

You're confusing the training method with the internal process. If I had you repeatedly attempt to learn how to make believable completions of partial documents about a given topic, you would eventually learn things about that topic and could use your knowledge to create more believable completions of documents about that topic.

Copper is expensive, and the manufacturing process sounds finicky.

It'd be interesting to try to imagine a Venusian colonization that's like two separated levels: an atmospheric area where most people live and where you grow food, and then a subterranean (yes, yes, sub... aphroditean?) where you mine and so forth, and there are brief, fraught transitions between the two layers but no actual habitation of the surface or lower atmosphere.

Not really. It's turn-based and hex-based.

People also do some really stupid things, I'll just throw out.

But it also kind of sounds like you're just saying that there's a scalar of "get a little better" before you'd trust AI with something.

Genuine question: is there a big inherent difference between "I don't understand this thing but I think this other human does," and "I don't understand this but I think this other AI does"?

If your answer is "yes," do you think that's inherent to the (metaphysical?) fact of it being AI or to specific limitations to current AI? If the latter, what changes to AI would let you trust it?

Quick heuristic for someone who claims that their AI is conscious: do they claim it's "a gender that they are not attracted to"?

Amazon isn't a monopoly, it has 8% of retail in the US.

There's no real evidence that this trade was made by Buffet himself, and it's part of a general major sell-off of Amazon that transparently did not temporally align with the idea that Amazon's retail business has suffered a decline in quality.

This is the market making a (reasonable!) judgment that it lacks confidence that Amazon's capital expenditures will pay off.

These just aren't the relevant concerns behind Amazon's stock performance in the last month. It's the capex.

Nonsense. There are lots of things that you need more than three strikes for, especially on a platform that you expect to use for decades.

I'm not here to say that Facebook's enforcement behavior is optimal, and I don't know that a "17 strike policy" is a full description of their enforcement behavior. But there are plenty of behaviors that you want to discourage but not go nuclear about.

No?

Like, I'd think that was a bad policy for murder in particular, but "we don't allow things but we give you a lot of chances to correct your behavior" is ordinary.

Native population is declining (and prime-age workforce is retiring), and the Trump admin has been extensively working to reduce the size of the immigrant workforce.

So the unemployment rate is staying low, but the absolute number of workers is flat or declining.

The in-industry terms for these are "brand marketing" and "performance marketing," FWIW. Brand marketing is the first thing, performance marketing is what you're calling sales.

Price discrimination is bad. It's worth trying to ban. You'll never stop 100% of it (and trying to go too over-the-top in terms of stopping it would not be worthwhile), but this is a useful area for regulation.

On some level the headline is like "yeah, no shit," but the surprising thing is the claimed strength of the effect. 50% absenteeism increase for missing a birthday congratulations? Really?

Apropos of nothing besides the mention of Disco Elysium, I present to you the best item in a CRPG: https://discoelysium.fandom.com/wiki/Volumetric_Shit_Compres...

The US has enriched a vastly larger number of software engineers through at will employment that Europe has through making it very hard to fire people who aren't adding value.