There is nothing simpler than *using* AI, just write a prompt and get your result. That's the whole point of AI, any dumbass can use it. "Prompt engineering" is overrated unless it is a system prompt of if you are doing stuff like jailbreaking or prompt injection, neither are particularly relevant to "normal" use, and not that hard anyways. The only thing you really need to know is that sometimes, LLMs hallucinate, so you need to verify. And sure enough fact checking is something all universities should teach, but it is not exclusive to AI by far.

Now *understanding* AI (i.e. machine learning) is another story. Algorithms, linear algebra, statistics, etc... These are hard to pass in many scientific fields nowadays, with so many papers involving machine learning. But it doesn't seem to be what it is about.

And BTW, I think one *should* ban chatbot use for classwork. Students will have more than enough occasions to use it outside of school. Instead, school should teach students how to think by themselves, so when they are inevitably going to use chatbots in other personal or professional situations, they can do it with a working brain. The "creative" ideas in the article are not, there is no creativity in having ideas being suggested by a LLM. "karma and the practice of returning shopping carts", come on, that's a meme.

UBI is what you get when you combine communist/socialist and capitalist ideals, if tried in full, it would be terrible.

UBI just gives a bunch of money to everyone, that, if spent wisely, will allow everyone to live decently. Except that not everyone spends money wisely, and I believe it contributes more to poverty than the lack of a job. Giving these people a bunch of money will not solve the problem. You can instead give them a house, food, health care, etc... but that's not a UBI, that's just regular socialism.

The whole point of UBI is to *replace* welfare systems, simplifying administration, making fraud irrelevant, etc...otherwise, that's just welfare, not a UBI.

None of the UBI experiments are true UBI, they are just increased welfare, because UBI simply won't work. How much welfare is right is the hottest question, I don't know if there is a right answer, but one thing is clear: UBI is not it.

There is a good chance that enshittification bothers him more than most, as he is the one to implement the technical solutions for all this shit, which is a lot of work that goes against making good quality software. And of course, if people start complaining about things like crashes, lag, etc... he will get the blame.

I guess he likes making money, so he is not completely innocent, but I am sure he'd rather make the money-making part somebody else's job.

BO went to space, the (arbitrary) limit for space is 100 km and they went over it. Going to space is about how high you go.

BO didn't get to orbit, but getting to orbit is not the same thing, it means going fast enough horizontally to avoid falling back on Earth, but not so fast as to escape Earth gravitational pull. Getting to orbit is about how fast you go. Some suborbital flights actually go higher than some orbital flights, for example, ICBMs. And we could imagine orbiting below 100km (so, officially not in space), but because of atmospheric drag, it probably wouldn't stay up there for long.

So BO went to space, but it didn't go to orbit. And yeah, orbit is much harder.

The article says "the edge of space" because usually, when people think about space, they think about at least orbital flights. Here it is like saying you went to Canada when you just crossed the border, took a picture and went back home. You did go to Canada, it is just not what people usually mean when they say, "I went to Canada" so you may want to use an euphemism instead.

The problem is that it is not really about pollution but about how new your car is.

Have an old, small car that is well maintained and doesn't pollute much, you are still banned, you can't go to a lab and show that your results are acceptable, go buy a new car. Have a new monster SUV that cheated its emission testing when it came to market, that's fine. No one is going to stick a probe down the tailpipe and test the actual emissions, at least not for the purpose of getting access to "low-emission" zones.

That's why the ruling is so unpopular, it hits the poor who can't afford new cars the hardest. It is even worse than the also unpopular "contrôle technique" which is a mandatory certificate for road-worthiness, to renew every 2 years. At least, the latter is based on actual testing, not just the age of the vehicle.

In education, it is a feature, not a bug. Research tends to consider that handwriting helps with retention *because* it is a slower and more involved process.

The problem is obvious: we want large screens, we want portable phones, tablets have large screen but are not very portable, non-foldable phones are portable but they have a small screen, foldable phones have both.

I especially like the Huawei tri-fold phone as it gives you a proper tablet-sized screen when unfolded, not an awkward square. At least in principle. In practice it is absurdly expensive and fragile.

If anything, tablets were more of a solution in search of a problem when they came out (if it doesn't fit in your pocket, you might as well buy a proper laptop) than foldable phones. In fact foldable phones were the most common form factor before smartphones, the reason it is not the case anymore is technical: foldable screens are expensive and fragile, solve the problem and I expect foldable phones to become the standard once again.

Or maybe inefficient in a way that can't get fixed by random budget cuts. We see a lot of inefficiencies in big companies that are profit driven and ruled by unelected, amoral and smart people who would do anything in the name of efficiency.

It may take 10 people to screw in a light bulb and an outsider will find it ridiculous, and sure, it is. However, they are all doing their part and if you fire just one of them, no more light bulb being screwed, the company will be in the dark and there will be trouble. So, not an option. Or, then, you can fire the entire light bulb screwing team and replace them with a single person and do like everywhere else where a single person can screw in a light bulb, except that the company uses custom light bulbs and no one person can screw it in. So you may want to replace your custom light bulbs with standard light bulbs, but doing so will have a cost, and for a time, and maybe forever, you will have both the 10 people team for the custom light bulbs and one more guy for the standard light bulbs. So, 11 people to screw-in a light bulb...

Or you can just fire these guys anyways and let the company go dark. That's essentially what Elon Musk did to Twitter/X, that indeed was way overstaffed for what it delivered. The result: it is still overstaffed, even if less so, but now, it also runs like crap now and they are losing users and customers. Maybe it will recover and maybe it will run more efficiently later but it is not a given. The difference between Twitter and government services is that we can do without Twitter, but we can't do without government services.

There is a saying that goes "for every complex problem there is an answer that is clear, simple, and wrong", DOGE is such an answer.

Trump is exploiting a loophole. He cannot impose tariffs unilaterally unless there is a threat to national security. So everything is a matter of national security.

I am all for making CS (actual computer science!) a requirement. But AI, nah, except maybe as a side note about what it *cannot* do, or maybe in the context of linear algebra.

The reason I think teaching CS and coding is a good thing is for the same reason why teaching maths above simple arithmetic is a good thing: it teaches logical thinking and a rigorous approach to solving problems. When you write a computer program, you can't be sloppy, what you write is what the computer will do, the computer will not do what you think you have written. In a world where we are surrounded by machines, I think this is an important skill to have, machines don't think, you have to do the thinking for them, and programming puts you right mindset. It can even extend to giving instructions to people, contracts, etc... especially in an industrial world where you and the person making your stuff will likely never meet, and all they have is what you have written.

  In addition, coding is a useful skill by itself, and it can be fun because you get concrete results.

AI is totally unlike that. The core of it is maths, but that's maths that is already taught at school, at least in STEM fields, so nothing special here. There is linear algebra, derivatives, standard stuff, math classes can be updated to talk about this application but that's about it. But beyond that, it is sloppy and doesn't require much skill, in fact, that's the whole point, and even the experts don't really understand how it works besides the fundamental maths. So what there is to teach at school (beside maths)? In practice, it could be like sexual education. We don't teach kids how to fuck, they don't need classes for that, but we teach them biology, and also the risks (which in case of AI would be things like false information/hallucinations).

The be fair to these posts, BTC eventually reached $100k and on average for the last 5 years progressed a lot more than the global market. So maybe these predictions were optimistic on the time frame, but I am not going to call "LOL!". Baseless or not, those who bought based on these predictions made money.

I am just looking at the charts, I have no skin in this game and I don't intend to. Will it exceed $250k one day? Maybe, maybe not, maybe it will crash, but $250k in a few years is not out of the question. Very improbable for the next month or so, but If I was sure it would get to $250k in the next 5 years, it would still be a very good investment.

I don't like gambling, I don't intend to play Bitcoin, and I don't have any regrets, but I must admit that some people have hit the jackpot, and my reactions is more "GG" than "LOL".

Yoda's specie is made deliberately mysterious, but Georges Lucas supposedly said that Yoda is "actually the illegitimate child of Kermit the Frog and Miss Piggy."

That would make him only half pig.

For the transphobic part, I don't know, but it may have to do with his frog side, as some frogs are able to switch sex. Maybe, as a hybrid, he lost this ability.

So, while there is some merit to it, I don't think that "transphobic pig" is the right term. "Gonochoristic frog-pig hybrid" would be a more scientifically accurate description.

Maybe they didn't need a scene that is more than 2 minutes long, and that the actor getting a sore arm from time to time and the occasional blooper didn't justify the added complexity, cost, and loss of flexibility a support rig could result in.

I agree that it is most likely deliberate.

I have a friend who is French and living in the US as an actor / voice actor. More often than not, he is asked to keep his French accent, sometimes to the point of overacting. And it makes sense, they cast a French guy to play a French character with a French accent.

For Mr Rutte, I believe it is the same idea, he is a Dutch politician and even as a secretary general of NATO, he still represents his country, and his accent is a way to insist on that part. He is not American, he is not British, he is Dutch and his country matters in the international community. I think that's the message.

Made up expressions and etymologies happen all the time. Like made up quotes from famous people.

When I see such explanations online, or by word of mouth, I usually assume it is wrong, and it usually is. Even generally reliable sources (ex: Wikipedia, major news outlets, etc...) often get it wrong, sometimes you need PhD levels of research to get to the bottom of it. Of course I don't go that far, I just treat these "explanations" as good stories, not facts I can rely on.

That LLMs make up good stories too is fitting. And if the story is good enough, a new expression may just have come into existence. Language is not static.