expr

1. Supposedly there's a way to install nix without root access, but I can't speak to it as I've never tried. Ofc it doesn't require sudo to install packages or anything, though.
2. I don't think it does this right now, largely because it's super fucking complicated (as is basically everything Apple) and homebrew casks themselves have had a ton of headaches around it. But nevertheless, I think home-manager has some workarounds it uses itself to enable many common GUI apps on MacOS.
3. Not sure exactly what you mean, but I think it does that?

If you want to install packages purely by name, you can use nix-env -i hello or whatever. But it's pretty janky and not really a recommended way of doing things.

Nix. I use it for everything, including all of my tools I use on my work MacBook.

There are many ways to use nix for this stuff, but personally I use home-manager in a flake-based setup. Versions of tools are all pinned in a lockfile which is committed to source control, so it's easy to get my config and all my tools on a new machine without any breakage (it does require installing first, though).

It's a great tool and has largely solved the pain of dealing with having to work on MacOS, for me.

Have you used Jira? It's a memory guzzler

Kind of, though they honestly just do pretend immutability. Object references are still copied everywhere.

It has to do with countably infinite sets.

The analysis on Wikipedia does a better job of explaining the concept: https://en.m.wikipedia.org/wiki/Hilbert%27s_paradox_of_the_Grand_Hotel#Analysis

The whole point is that it's something we can prove mathematically that is highly unintuitive.

Forced to use copilot? Wtf?

I would quit, immediately.

The writing in the second season was much worse, too, unfortunately.

As much as it pains me to say it as an American, you're in danger from the US, now. It is now a fascist dictatorship with imperialist goals.

Admittedly I haven't worked on any games, but if I were to do so, I always believed ECS to be the way to go: https://en.m.wikipedia.org/wiki/Entity_component_system.

Yep, senior Haskell developer here and I have had their recruiters hounding me many times, even though I have told them to fuck off again and again.

I always find it so funny that they chose Haskell. They are desperate to hire, but no one in the Haskell community actually wants to work for them. I'm in a discord server with a bunch of veteran Haskellers and everyone there won't touch them with a 100ft pole.

You can read the full paper yourself here: https://arxiv.org/pdf/2501.02305.

I haven't had time to fully read it yet, but glancing through, it looks pretty legit.

This is a graduate computer science student working with accomplished CS faculty at Rutgers and Carnegie Mellon, we aren't talking about some rando making outlandish claims.

The thing about theoretical computer science is that, like math, it isn't subject to the pitfalls of empirical science. It isn't dependent on reproduction. The proof is provided in the paper, so either it indeed proves what it claims to, or the proof is erroneous, which can readily be refuted.

Did you read the article? The claim is that they have invented a new kind of hash table that has vastly improved algorithmic complexity compared to standard hash tables.

I haven't read the paper yet, but if what the article claims is true, it could be revolutionary in computer science and open up a ton of doors.