hyperpape

I haven't previously thought about this, but I think words over a commutative monoid are equivalent to a vector of non-negative integers, at which point you have vector addition systems, and I believe those are decidable, though still computationally incredibly hard: https://www.quantamagazine.org/an-easy-sounding-problem-yiel....

With all due respect, this is completely wrong.

There is a difference that someone smoking nearby automatically harms people around you. With alcohol, the effect is more unpredictable, but it is equally real.

Alcohol is a factor in an automobile crashes, and a factor in a significant proportion of violent crime, especially domestic violence (https://www.cato-unbound.org/2008/09/17/mark-kleiman/taxatio... edit: this source isn't as great, Kleiman has written elsewhere about the subject, but google is failing me). If we could wave a magic wand and cause drinking to cease to exist, many lives would be saved.

Note: I do in fact drink, I am not a teetotaler. But what I said above is factual. I personally believe that prohibition would be worse, and it's reasonable for individuals to make their own choices. But that does not entail denying that it goes very badly for many.

This is a great piece of data, but only a piece of the actual question that we need to answer, which is:

For a given input, how many tokens will be used for an answer, and how high quality will that answer be?

Measuring the tokenizer is just one input into the cost-benefit tradeoff.

This is an interesting analysis, but "are the costs of AI agents also rising exponentially is?" is a very bad question that this doesn't answer.

What's rising exponentially is the price of the most ambitious thing cutting edge agents can do.

But to answer whether the cost of AI agents is rising in general, you would take a fixed set of problems, and for each of them, ask "once it's solvable, how does the price change?"

For that latter question, there isn't a lot of data in these charts because there aren't enough curves for models of the same family over time, but it does look like there are a number of points where newer models solve the same problems at lower prices. Look at GPT5 vs. the older GPT models--the curve for GPT5 is shifted left.

His book appears to have been published in 1942 or earlier: https://time.com/archive/6786636/books-biography-in-pictures....

Thanks, this is a good explanation, though I would not have phrased it the way he did.

> Founders who started pre-2025 typically have built a technical stack optimized for a world where software development was bespoke and expensive.

Of all the things that AI has changed, tech stacks aren't one of them. The bots will gladly write Typescript, Java, Python, Rust, what have you. They could not give less of a shit.

As blind as my belief that Asia exists, because I haven't personally navigated there. Hell, I've used electricity (using it right now), but I couldn't do the experiments you need to do to get myself to an 1850s level of understanding of how it works, much less our current level.

I trust that Linux has a process. I do not believe it is perfect. But it gives me a better assurance than downloading random packages from PyPi (though I believe that the most recent release of any random package on PyPi is still more likely safe than not--it's just a numbers game).

No better proof that they're a huge company than that I could forget about an entire public cloud offering. Good point.

1. They maintain and sell one of the largest relational databases.

2. They're the primary maintainer of one of the largest programming languages.

3. They do tons of HR/ERP type software.

4. They have a supply chain division (my company is a direct competitor, and we have 2000 employees--it's a drop in the bucket, but a few thousand here, a few thousand there and it starts to add up. Afaik, their supply chain org is bigger than ours).

5. Other things I probably don't know about.

Many of these things come with swarms of consultants who implement the software for companies that don't have any internal technical competency, which swells the number of workers by a lot.

Don't get me wrong, I'm not remotely a fan, I like to quote Bryan Cantrill's rant. However, they do a lot of things.

If I do my python right, from 2010-2020 they grew by 2.5% annually, from 2020 to 2025, they grew headcount by 3.7% annually.

After the layoffs, they'll apparently now have grown by 1.0% annually since 2020.

So yes, from 2021 to 2023, they had a huge spike, but overall, it's a net slowdown in growth relative to the 2010-2020 period.

If this was about reversion to the old pattern they'd have done a smaller set of layoffs or simply wait for a few years of zero growth.

Why didn't you use log-scale? It seems like the obvious call.

> Magnificent 7 companies are increasing capex to their biggest ever to differentiate their tech from each other and the big AI labs, but the key realization is that they don’t have to spend it to win. It’s a defensive move for them, if they commit $50B, OpenAI and Anthropic need to go raise $100B each to stay competitive, which makes them reliant on investors’ money.

Stay competitive how? If the Magnificent 7 aren't spending the money, then how could it possibly hurt OpenAI/Anthropic to not raise equal amounts of money? Maybe you can pull together an explanation, but this author didn't even try to do so.

This piece seems poorly thought-out, but well designed to get shared.

Promote writers who will actually explain their claims carefully.

Returning no results is going to be linear in any DFA or NFA based implementation, though. You go character by character, and confirm that there are no matches.

It's only when you return multiple matches that the engines have a problem and become superlinear.

Given the subject, it is funny to me that this post is meandering and repetitive.

Most of those meetings are pretty damn fluffy. No one goes back to their desk and does anything different because they've introduced new company values and the acronym is S.M.I.L.E.

But this meeting is a course correction for how they're using AI, which is a huge initiative. He'll be trying to sell the right balance of "keep using the technology, but don't fuck anything up."

Too cautious, everyone freezes and there's a slowdown[0]. Too soft, everyone thinks it's "another empty warning not to fuck up" and they go right back to fucking everything up because the real message was "don't you dare slow down." After the talk, people will have conversations about "what did they really mean?"

[0] If you hate AI, feel free to flip the direction of the effect.

Right, but if you say something essential in a meeting with 10 people and it has to percolate through five levels of management to reach the front-lines and gets watered down, that could be much more lost, even millions.

Scale cuts both ways.

What matters isn't how big the meeting is, it's how important the material is, and how well presented it is.

I think the most significant boundary is given by the question: "is there a plan to support new minor versions of Python?" It sounds like there is not.

There may be non-zero maintenance work happening, but a project that only maintains support for old versions and will never adopt new ones is functionally one that the ecosystem will eventually forget about. Maybe you call that "under active development" but my response is "ok, then I don't care whether it's under active development, I (and 99.9% of other people) should care about whether it's going to support new minor versions."

On the other hand, if you don't support new minor versions day one, but you eventually support them, that's quite different.

One of the most praised recent restaurants in the United States is based on an attempt to reconstruct pre-Colombian cuisine from the Americas: https://owamni.com/, https://www.newyorker.com/magazine/2022/09/19/how-owamni-bec....

Your hypothesis: you and I don't have a good enough version of Claude. My hypothesis: you and I don't have missiles.

Good observation, I meant at the border. Your passport won't matter that much if you get stopped by a cop.

But also, look carefully at the comparisons I offered. I didn't include all the combinations, because I only was including comparisons that were obviously true without any room for ambiguity or nitpicking.

As you noted, a black citizen might be treated better at the border and worse during a traffic stop compared to a white foreigner.

Yes, but...

The health of the market is not a function of the total number of jobs alone, it's a function of the number of jobs and the number of people to fill them.

The number of total jobs going up year after year meant that there were increasing numbers of candidates, new people entering the field. If the job growth stops, then there still we be candidates coming in. There will also be the new hires from the last decade moving into increasingly senior roles, and there won't be space for them (unless you devalue the meaning of "senior" even more).

So the year over year change matters a lot. If it plateaus, or even declines slightly, it's more than enough to make a terrible market.

1. A few messages upthread, I note that I'm an American, and I'm from the South. Quite familiar with how racist folks can be.

2. Reread what I wrote, it's not contradicted by what you said.

hell of a typo in that last sentence, and I can't edit.

I meant to say "doesn't mean you're not less likely to be targeted".

Don't put words in my mouth, don't say silly things.

I'm well aware the color your skin matters a lot, but your passport also matters, especially at the border.

You're better off with white skin and a US passport than with white skin and a British passport, but you're also better off with brown skin and a US passport than brown skin and a British passport and that's still better than brown skin and a third-world passport.

And yeah, even if you're a white man with a US passport, you still might end up shot by ICE if you're in Minneapolis (doesn't mean you're less likely to be targeted).

I am a US citizen living in Portugal. I have the right to go to the US, live there, etc.

I recently went back for a funeral, and I had to spend a moment reminding myself that it would be fine for me.

For people who don't have my passport, I wouldn't feel comfortable telling them "it will be fine", though I would still tell a European "the odds of a problem are relatively low." But I couldn't in all honesty say "there's nothing to worry about."

> If Walmart can forcibly add verbiage banning AWS from it's vendors and suppliers, the US government absolutely can.

You can add new language to new contracts. That is not what this is.

Although, when I learned foundations of mathematics, every function was a set, and if you wanted them, you'd get plenty of junk theorems from that fact.

Go is also winner take all. It's psychologically satisfying to have a big win, in the same way that it's psychologically satisfying to achieve a brilliant checkmate, but in any ordinary game or tournament (outside of certain gambling setups), a win by 1/2 point is the same as a win by 20+ points.

Ah, so it's free, but you still have to wait 3 months. Just a question...what are you waiting for?

Of course the answer is all the things that aren't free, refinement, testing, bug fixes, etc, like the parent post and the article suggested.