Cleaning up after a child is not the same as general cleaning.

If you’re an adult, you can get away with daily dishes, and ither cleaning once a week, because you’re capable of taking food all the way to your mouth without dropping it, and you probably have a vague laundry organization system, and some of the time, you out stuff back where you foundil it when you’re done using it.

Kids cannot do any of that shit, so your house accumulates one week worth of responsible adult mess in about 11 minutes.

I have a confession:

When I joined, I thought ML was for Machine Learning.

Shit.

That’s super dark and super dystopian… and I feel terrible for chuckling.

A shame that John C Woods isn’t still around.

I dunno, we already have fast chargers that can go from 20% to 80% in 20 minutes. A full tank of gas still takes 5 to 15 minutes (big tanks).

I think it’s feasible that we could someday figure out chargers that are even faster than standard pumps, for a comparable mileage.

I know I’m being silly, but my brain likes numbers like 5 and 10. I can’t imagine saying 10 to 13. And a 5 degree swing in C feels too big.

Yeah, but that’s because that’s what you’re used to with the F scale.

If you were estimating a number of tablespoons of sugar, you wouldn’t be estimating in groups of 5, because that’s crazy. You’re probably more accustomed to estimating one tablespoon at a time. The same with estimating measuring cups of milk.

Or, if you don’t cook very much, for estimating walking distances, you’re not estimating in groups of 5 miles. You might group by quarter miles and say, “It’s a quarter mile away,” or a half or something.

There’s nothing special about this scale. All of the supposed benefits are the result of your familiarity with it. Had we been using °C the whole time instead, you would be estimating degrees two at a time instead of 5, and everything would be completely normal.

Celsius and L/100km are not useful or show it’s not a great system for everyday use.

This is probably just a matter of taste and convention. If everyone knew what 12°C was, then it would be perfectly fine for everyday use. Everywhere else in the world, people are using it for everyday use, and they’re doing just fine.

Same for L/100km. Oops, you specifically wanted to know how far you were going to go with 1 Liter? You can convert that just fine. But it’s just as likely that you’re looking at mpg, and you actually wanted to know how many gallons it would take for you to go from A to B, in which case you also need to do a conversion.

Exercise:

• Set a timer for 1 minute.

• Sit down and do nothing.

Do that every day. After a while, you’ll adjust and it wont be painful. Then do 2 minutes. Then do 5. Then do 15. Every day.

Here’s what happened:

You’ve lived your entire life in a bowl of cocaine. You were born into it. When your parents wanted a break, they gave you some coke. When you were good, they gave you some coke. After you completed your homework, you were allowed a little coke. When you grew up, if you feel tired, or feel like you don’t want to do something, you can have a little nose candy to recharge.

If you want to stop using so much cocaine, you can’t just go cold turkey. You need to start by being able to survive one minute at a time without it.

It’s easy to just remove the bowl of cocaine and not use cocaine. But that’s not going to help you practice choosing not to do cocaine when it is available.

LLMs are terrible at math.

Mathematician chiming in: they’re not nearly as bad as they used to be. Straight up to differential equations and linear algebra, they’re pretty accurate. The writing is on the wall for some serious problems in the very near future, because basically every student in college right now is just offloading their math abilities to the machines.

Some of the seniors are good. Most of the rest of them are not.

It might even be one of those things that they can’t fix, simply due to how LLMs operate.

For high-level math, there are some flaws, yes. They often come up with wrong answers if you ask questions about problems that there aren’t very many sources on. There are hundreds of thousands, and maybe millions of differential equations papers… but there are only tens of thousands of Algebraic Topology papers. It’s simply a smaller training set.

However, they’re still giving right answers most of the time. I’ve only had one situation where someone in a group said “bUt cHatGPT sAiD…” and I had to give them the bad news that it was wrong.

In any case, we’ve had calculators for a long time, yet children are still taught math in school because society recognizes that understanding as crucial.

I think that the fundamental difference is that the tools replace different things. A calculator makes it so that you don’t need to algorithmically process 5 sig-fig numbers without error, several times per problem, in order to have your answer be “correct”. You still have to understand that you do, in fact, need to multiple the length of the field by the width to find the area of a rectangular field.

LLMs are replacing the need to actually understand anything about how a rectangle works.

Absolutely.

Fundamentally, only reading about ideas that you already agree with is bad for you. You will learn less about your beliefs than if you had read something that challenges your beliefs.

Let’s say you’re very pro-capitalist, and you want to know everything about capitalism, so you read Adam Smith’s Wealth of Nations. Great! You learned some things about the historical basis of your beliefs, and there’s value in that. So then you read Capitalism and Freedom by Milton Friedman. Okay! So you’re tripling down on the same ideas. You probably already know most of those ideas, though, right? There will be some new stuff, but ultimately it’s more of the same stuff you already know.

So you try some more recent authors, and give a shrug to Ayn Rand, and you give Lions and Scavengers a shot. And maybe you pick up a new thing or two. But mostly, it’s the same ideas that you’re already familiar with.

Do you know what none of these books have? An angle of critique (except Smith, sort of). They’re all just listing all of the good things about capitalism, and they’re not telling you the bad things. Do you believe that Capitalism is a perfect system to organize a nation’s economic policy, with no flaws? No? Then you should be interested in someone who elaborates on some of the negative aspects of capitalism, and what can be done to address those flaws, right? Because if there are flaws, you want to know what they are, so they can be fixed, right?

That’s why you should read Das Kapital.

Factually, using AI in your education means you get less value out of that education. In every math class ever, you’ve probably heard some version of,

“I don’t care that you got the right answer. I already know that 2x^2 - 5x -3 factors into (2x + 1)(x - 3) and I already know that this information is freely available. What I care about is making sure that you can figure that out without anyone helping you.”

If you use AI to do a problem, you’re training your ability to do something that everyone else on the planet can do. If you force yourself to do it, you’re training your ability to actually think.

You actually WON’T have to do more work in the long run. Think of it as an investment. If you spend the time learning it now, properly, so that you actually understand it, then every test that your study for, for the rest of your life, will be easier, and you’ll get higher grades. Measurably.

https://www.media.mit.edu/publications/your-brain-on-chatgpt/

https://www.media.mit.edu/articles/ai-s-impact-on-cognitive-ability-mit-study-reveals-more-troubling-data/

Keep it up.

…

I’m going to acknowledge that you responded so that, when I don’t continue this conversation, you’ll know that it wasn’t because I decided to abandon my position.

You could say your response did not have a significant impact.

Did anyone do studies on the homework and exam scores of people using compilers vs people not using compilers?

Did anyone establish that people who used compilers are measurably dumber than the people who didn’t?

No?

Well then obviously it’s not the same situation, now is it?

It’s like you didn’t read the article.

You shouldn’t have to take a class for an operating system…

…So we should use android? Maybe chromeos?

Invalid inference.

Well, there’s fairphone. And farmers markets.

Honestly, most of the things people buy are things they don’t even need… housing, cars, and insurance excluded.

Bottled Water by Nestle? ❌

Learning to love the taste of your local tap water? ✅

Now that’s what I call Ethos!

I think most of us are on-board with police body cameras. The fundamental difference is that they are recording things in the presence of the police, and that they have the measurable benefit of holding the police accountable to their actions (sometimes).

Rats are good people

$60 per kilo

That’s not bad for a prototype! With lab-growth meat, the initial cost was $2.3 million / kg. It’s come down since then because of process and scale improvements, but when it starts off that high, it’s going to be a while for it to become affordable.

A quick non-AI figure is $0.60 per kilo to recycle plastic… so we just need improvement by a factor of 100 which is quite realistic.