You probably know this, but a note for the benefit of people who don’t. The entire git history, including metadata, can be modified. Unless you have an independent offline remote to compare to, this method is not 100% guaranteed to detect tampering in all cases, for example if the nixpkgs repo is compromised (or your machines’ connection to your git forge is being MITM’d)

The success rate of computers doing those tasks has gone from 0% to 70%-90%, but reaching 100% might take a very long time.

How many people are needed to make up the difference to 100% though?

You and GP must live in the Bay area. Not once in my life have I seen someone coding on a plane, let alone next to me.

I’m east coast but I just use LSP and a local environment - no need for WiFi or anything like that to get work (or fun) done

Nope. The flight was from California (not the Bay) to a popular resort destination.

I consider Azure risky and I would love to hear more about the companies and sectors that would laugh (with me). So which ones are you talking about?

LLMs will often helpfully predict made up tokens for the content of the data fields.

For 100% of jq use cases I have the data wouldn’t fit into context. But even for the smaller things, I have never, not even once, had an LLM not mangle data that is fed into it.

Take a feed of blog posts (and select the first 50 or so just to give the model a fighting chance). I’ll give you 80% likelihood of the output being invalid JSON. And if you manage to get valid JSON out of it, the actual dates, times and text content will have changed.

I’ll have to give this a shot.

One possibility: Claude Code subagents get their own 1 million token context window; should be better with large JSON files vs. having everything in the same context window.

Human brains might work as lower power consumption, higher efficiency cores for less demanding work.

Ukraine draft age limit is now 25. It was lowered from 27 in 2024. If you are young and heading to university, you’re safe (for now). Even if the age limit is lowered, it will likely apply to NEETs first.

Good to know, but my comment is not about Ukraine or any other country in particular

When people say ”time is money”, there is usually an amount of money that makes you go ”… but then again a lot of money is also money” and give up some time in exchange for that money.

Wartime resources are similar, in the sense that they’re at least partially fungible. Maybe the amount of money the Gulf states are willing to pay for borrowing those people buys you a lot of the other stuff.

It’s sad that it’s come to this, but your addendum with credentials and a non-slop declaration worked. Three years ago it would have seemed unnecessarily self-aggrandizing. Today I wouldn’t have clicked the link without it.

I would not assume that giving my kids $400 million would be a net benefit to them.

Now to be fair I might be wrong, since I’ve neither researched this nor given it much thought. Maybe there is research on deca- and centimillionaire heirs that shows positive effects of money on life satisfaction, happiness, health and other life outcomes. However I suspect it works similarly to sheltering kids from adversity, failure and hardship in general: disadvantages them psychologically and leads to more problems down the line.