I tried the way they used "oracles" to verify their implementation against known good ones, and I must say: this works.

Not all implementations are re-implementations, so this approach won't work for everything. But for a new implementation in a new programming language than the original implementation, it works great. Built myself a MIB compiler, checked against smidump. Now it is more correct than the original, because smidump still crashes on some inputs, while mine does not.

So the news is not so much that Anthropic built a C compiler, but HOW.

While they are not as efficient or flexible, they are many times more efficient than resistive electric water heaters. I've installed one with in house air intake (due to construction reasons) in my house and it cooled down the basement by a few degrees (and removed air moisture as an added bonus). In summer the thermal capacity of the ground heats up the basement again, in winter it's a bit cooler, but it still works efficiently.

They are efficiënt but all that efficiency is eaten up by electricity being more expensive than gas.

Not sure how to take this. While your statements are objectively true, there are a lot of reasons the US won't attack the EU, even if the reasons are mainly economic.

Russia on the other hand is shit at waging their war in Ukraine but they achieved their objectives (land bridge to Sevastopol, etc.).

So the worry is not that Russia will wage a well fought war, but some war at all, even if they are shit at it, because it will do extensive damage either way. And we know for a fact that they don't shy away from it.

Now make an algebra out of the CAP theorem. It's not already one, isn't it? Didn't read the paper.

Newer wind turbines don't have a gearbox and are almost completely silent. When standing next to one, the loudest components are the electrical inverters/transformers.

Wings are loud

The API is great. Will definitely try it out. I have a use case already. How difficult would it be to extend this to support timed flushes? Like, every 200ms or so, regardless the fill of the buffer?

That's what the `window` option does (https://github.com/hackermondev/batched?tab=readme-ov-file#b...). After the first item is added to the buffer, it waits for the maximum window time, then calls the inner function with the buffer fill.

My hunch: it's not a real captcha on their page femboy.cat, but actually a script which "claims" the address in the ipv4.games game. Nothing to see here, move along.

Nothing so complicated. The HTML source of femboy.ca/ literally contains:

    <img src="https:////ipv4.games/claim?name=femboy.cat" hidden>

However, the question is why would this domain get 20 million distinct visitors (before being posted on Hacker News)?

Interesting. I looked up top 10 domains. I wonder if you could have it occur on wikipedia.org? Other random thoughts after looking at https://radar.cloudflare.com/domains... Maybe tiktok somehow? Or a bug in NTP? :D

For thorny problems I let the agent give me a simplified flow-chart in mermaid syntax. LLM's brain-farts are easily visible then. I correct the flow-chart "Ah, you're right!" and then let it translate it to code. Works wonders.

I often provide mermaid diagrams in my promt. Mermaid seems to be a good common markup to communicate relationships between humans and LLMs.

that's smart! I’ve actually been thinking about integrating something like Mermaid flowcharts directly into Nia’s output—visual context can make cursor etc understand context way better. have you found any particular types of problems where the flowchart approach really shines (or falls short)? Would love to hear more

That page can't be very old because it doesn't use any frames!

Well I've seen worse bookkeepers. "You know, you approved of the budget, but where are our customers payments in the balance sheets? We can't find them!" - "Uhm..."

Sure, let's compare an almost global failure mode with the worst examples of knowledge workers. ffs.

I’m coming back from my accountant right now. He uses winbooks and has interns doing the books. I have no tooling to do it for him, and I’m seeing absurdities such as a 4000 usd refund being processed as cashback, and simple typos not being caught. I wish I was working with an AI instead of this nonsense. It’d be far more accurate.