titzer

We forgot to run it in docker. Oopsie

Before LLMs took off I legitimately thought that most people would be paranoid enough to air-gap the first AIs that even had a remote chance of sentience so they didn't hack their way out.

Well, obviously not. You only live once! Just let the AI do whatever, I guess.

People are so unserious these days.

I feel like I've seen this movie.

I think over-complicated and over-engineered are not the same thing.

Over-complicated is adding too many features, too much mechanism, too many moving parts. Over-engineering is far exceeding the requirements in an unhelpful way. For example: build me a treehouse. Said treehouse could probably be made of wood. If you made it out of concrete and steel, it might be a heck of a lot stronger and last a heck of a lot longer, without being more complicated (just more expensive). Too strong is over-engineered. If you made a treehouse with 13 bedrooms and an elevator, glass windows, solar power, and running water, that's overcomplicated.

There was a long concerted effort to disable all the accountability mechanisms. A massive failure across the board, at every level, in every party.

Now that the entire justice system is broken, the American system is being dismantled and looted.

2024 was fatal. Enjoy the slide into the abyss.

The primary check on government growth

Disagree. The primary check on the government is consequences for violating the law and the constitution. That's just all gone out the window.

Power without oversight, no matter how distributed, will tend towards either complete chaos or tyranny (or both, if such a thing is possible). Giving three mutually distrustful criminals guns will not cause law and order to spring out of nothing.

An informed, angry and unforgiving citizenry is the only check on government, and we ain't go that.

Hard disagree. Public funds are absolutely for funding research into things that affect the public on a large scale. That's the whole point. What could be more "general welfare", as envisioned in the constitution, than making sure we are not screwing up our collective home?

Ah yes, we can't trust that our elected officials understand their duty well enough just fund Science and find out stuff works, so better throw up our hands and let the "market" do it.

Your view isn't controversial because it's daring, it's just plain nihilistic. It's just anti-government dogma which is cultivated by an incredibly cynical media atmosphere.

The media in the US is utterly feckless and broken.

Yes, now we can do thinking without thinking. Good job.

AIs should be forced to show their work. Every tool they use, every program they generate and run in the background, and every logical inference. They should be forced to produce Lean or Rocq proofs or execution traces for all the computation they use. For facts, they should be able to produce sources. For any abstract reasoning, they should be able to break it down into explainable steps.

Then, on top of that, they should be able to explain any of that, at any level of detail, whether talking to an expert or a layperson.

When satellites smash into each other at high velocity, they explode. Some of that debris will end up in higher orbits and linger.

I've picked up enough plastics off the beaches of the world already. Plastic ruins ecosystems and is a scourge on this planet. It's too late to fix, it's everywhere. 8 million tonnes of plastics enter the oceans every year, and it just keeps accumulating. We're ruining marine ecosystems.

Typechecking is more less picking a decidable set of invalid programs and making an efficient checker for them. As for ruling out programming styles, a language is, after all, Turing complete, so one could write an IR and an interpreter and then encode the rest of the program in the IR...

Personally, while I see the social advantages of having all code in one language be similarly structured and use the same idioms, I think a language is meant to be used, and it should offer all the building blocks for writing programs in whatever is the most natural way.

Because rich people want more money. They keep saying it's for the good of humanity and some useful idiots keep repeating that. A more balanced tax policy, regulation of financial markets, incentives to renewals, stopping investment in fossil fuels, increasing energy efficiency, and rewilding could all help, but we won't do a lick of it, because every single one of those things goes against the billionaires who own the political class.

The oldest DVD I own is from 1996 and it still works without any problem. Take care of them and don't let them get scratched up. Rip them and make a backup.

Knowing is input and skills are output. If you can't produce some kind of output in response to input, you've got a nice imagination but no skills.

8 hours in fuzzy spot one, 8 hours in fuzzy spot two, fitting some gaussian distribution, and then a few points at other times, and boom, not anonymous at all.

I didn't read the law but the article referred to it as "personal data". It's important to note that Big Tech believes it can "anonymize" fine-grained location data but just stripping PII and assigning a semi-stable random ID. They'll then argue this law doesn't apply because of that.

That's absolute crap because geolocation data is extremely easy to de-anonymize. I wonder whose phone is at my house 12am-8am everyday, takes more or less the same route through the city, and spends hours in my office everyday. I wonder.

This just makes reviewers and authors lazier.

The purpose of code review is multi-faceted. Hard to maintain? Yes. Might have bugs? Yes. Can be done simpler/cleaner? Yes. Is in line with project code style? Yes. Get someone else to also understand the code? Yes. Onboard junior team member? Yes. Sanity check design decisions? Yes.

This flippant note is mostly more self-justification for being a lazy code reviewer.

It's even worse when Google believes they have a legally defensible justification that your data has been "anonymized". E.g. "anonymized" location data directly from your phone that just so happens to be accurate to the meter. Such data just cannot be anonymized.

Or buying some crappy app off the app store, from which they take a cut.

By analogy, would complaining about any organization ridiculously more powerful than you (e.g. a government) without having a complete alternative ready to go also be "childish"?

People are going to come out of the woodwork to argue about OO specifically, and I'd probably join in, but the higher-level view that I've come to over the years is that one programming paradigm is not the best in all situations. It really helps to work on a lot of different systems that are not all in the same class and look at a lot of code written by people of all skill levels. It's not possible to really talk about all programs, but I will say that the OO way of thinking can very easily be overdone and end up a giant mess. Similarly, the FP way of thinking can also be very easily be overdone and end up a giant mess. The old-style procedural programming (C with a pile of global variables) can easily be overdone and end up a giant mess. Yet all three of those can actually work out great at the right scale on the right problem!

I think the problem is that we can't really architect a program well until after it works and fulfills all of its design requirements. Those are usually in flux and then we have to engineer dimensions of extensibility to prevent committing to the wrong architecture in the beginning. Those extra dimensions of extensibility come at some cost too, as they often mean indirection and cognitive overhead.

We often don't write the same program twice, or even three times, unless it's a really important program or a we just like doing it. Contrast to other fields, e.g. building houses, where we have lots and lots of examples of how to do it and they started to form archetypes. We try to call those design patterns in programming, but they're actually pretty vague and low-level. Can you imagine going to build a house and thinking "yeah, I think I need to use the wood-and-nails design pattern here, and I think I'll do rafters-and-shingles there, with some brick-and-mortar there." Yeah, real specific ideas on how to do it.

I have a fantasy of an alternate history where we as a society got our shit together and subsidized local libraries and ISPs so that they could offer cheap and even free NAS for everyone. Economies of scale and all that. It would have been a worthwhile public investment, but it's hard to justify spending public money on that and our politicians are so blinkered that cannot comprehend what "investment" means. Like...making it easier for kids to get smart? Why would we want that?

Instead we have the private marketplace fulfill all those needs for the low low price of ad infestation. Imagine how smart our kids would be if instead of 20 minutes of unsolicited ads a day, they saw 20 minutes of educational content and were required to pass a math quiz to access YouTube?

I have a large collection of DVDs that I've amassed over the years.

There's something nice about physical media; the bits are physically stamped into the medium. They're DVD-encrypted but I lawfully extract these bits and view them regularly.

When streaming services start on-the-fly editing for content[1] and revoking licenses, they can absolutely shove it up their butts. My old man take is that if a TV show or movie or whatever isn't worth putting onto a physical medium and distributing it to people who will buy it, I won't miss it if--I mean when[2]--it's gone. I mean, these huge movie studios act like pirates are going to ruin their massive profits, when they won't.

[1] And yes, they will absolutely on-the-fly, 1984-style edit films and TV shows for content.

[2] And it will go [https://en.wikipedia.org/wiki/Digital_dark_age] along with lots of other things into the memory hole of the digital dark age.

I don't see any data for anything past 2019.

Second greatest.

Well don't accept code from anyone ever then.

But seriously, if your format requires extensibility to the point that it embeds a bytecode, especially a Turing-complete bytecode, what format are you going to choose? Just design a new one? That's how you end up with a scripting engine with three ROP exploits.

But Wasm has deterministic execution, so if decode fails for you, it should have failed for them. I.e. it's not a problem that your system has introduced; they should be able to reproduce the failure independent of any client.