If an ad is funny, it's entertainment!

If the world achieves balanced trade by equalizing "imports per person" between any two countries, then the most populous country in the world will have a trade deficit, since it will import more than it exports from any other country. And the least populous country will have a trade surplus because it will import less than it exports from any other country. I'm not sure what this would mean or if it is even possible.

Let's assume that there are only two countries in the world, Canada and USA. Equalizing the imports per person at let's say $1000 per person per year, would mean that Canada should import $40 billion from the USA and USA should import $341 billion from Canada, per year.

Let's simplify it some more by replacing money with something tangible, let's say trees (a unit of lumber). If each person imports 1 tree per year from the other country, then Canada would import 40 million trees from USA and USA would import 341 million trees from Canada. Then do the same calculation for something else, let's say cars. Canada would import 40 million cars from USA and USA would import 341 million cars from Canada, one for each person. If I sit at the border line and watch 40 million things from USA exchanged with 341 million things from Canada, I might think, hey wait a minute, Canada is getting screwed here. They are giving 8.5 times more than they are receiving.

I conclude that equalizing imports per person is not a meaningful goal.

If he loses the case, the developer could name his thing "ECMA Bond".

"Script, JavaScript"

"Script, ECMAScript"

"Bond, ECMA Bond"

Can the AI also replicate the computer it's running on? Did they give it credit card and watched it as it ordered supplies, assembled a computer, installed itself on it and turned off the old system? We're probably not far from this scenario, but that it not even what they did.

If an AI system is capable of designing CPUs, GPUs, and AI techniques that are better than the existing ones, then I would call that closer to "replication" than what these researchers did. This may very well be done in the near future. Is designing a GPU harder than being the best chess player?

As I understand it, the researchers use "replication" to mean configuring a software application, by copying and adapting the configuration from somewhere else, including the model, which does not change at all, by using inference, instead of step-by-step instructions. In this case the application happens to be the AI server processes. I think it is an impressive accomplishment, but I wouldn't call it "replication".

In this case, meticulous shoppers could plan their shopping so the rounding is favorable more often than not. If, for example you tend to buy the same combinations of items often, you can adjust the amounts, so the rounding is down.

The store could always round-up and give you credit for the rounded amount somehow. There could be many interesting ways to do this. I'm not sure if it can be done without tracking you though. Some ideas:

A QR code in the receipt. Next time you shop at the same store (or perhaps any store), you could show them the previous receipt, and they could credit you for the previously rounded up amount.

An anonymous "change card". It would store round-off change, which you can use as partial payment in your next purchase at any store.

Airbus

quotes are from the article:

What is a particle in classical mechanics? An object with an exact location, zero volume, and therefore infinite density? Does this match what we experience in real-life? What is an object? A discontinuous distribution of matter in space, with an exact surface that separates matter from empty space? Those are obviously idealized constructs. A little experimentation shows that they do not exist. Solids break or melt, liquids evaporate, friction is everywhere around us. Real-life does not have perfect surfaces. Classical mechanics is a purely mathematical theory that happens to approximate real life under certain conditions. Didn't physicists question this impossibly idealized nature of classical mechanics even before they considered quantum mechanics? A wavefunction seems more real because it is continuous.

Measure position with infinite accuracy? How would you do that? What would it mean to know the position of something down to 1e-100 meters?

Quantum Mechanics was formulated not long after particles were discovered. Thompson discovered the Electron in 1897. Rutherford discovered the atomic nucleus in 1911, just 14 years before the 1925 formulation of Quantum mechanics. It was pretty obvious early on that classical mechanics could not describe atoms. So what did it describe? Idealized mathematical shapes with perfect boundaries and positions? Does this fit with what we experience in "real life"?