You don’t need a car in Singapore. Very good public transport and affordable taxis.

For more brain flipping try looking into hardware description languages (Verilog) or proof assistants (Coq).

Is this a new project that was intentionally started in PHP or something legacy? Any interesting benchmarks? Like minimal wire to wire network processing time and where the bottleneck is?

I saw md5 checksum implemented in scratch.

I would like to hear more

Canonical make it hard not to use snaps so only those who took extra steps are not using them.

Yea, not with firefox, at least not without switching to some third party repo.

Uses KDE and not available to general public.

If you open to do some research - you should be able to stay on Debian and use nix the package manager https://hachyderm.io/@fasterthanlime/111099224022408403

Comes with some steep learning curve, almost a learning cliff though. But once past this - it’s good.

That’s what I’m using, but it’s not a fully featured replacement.

Been using it since forever, no reasons to switch. It works. Got a bit upset at them when they killed xul/pentadactyl though.

Yea, something like that. Using it on my laptop already. configuration.nix for system plus home-manager for user stuff. Will move the desktop soon-ish.

NixOS. I’m going to migrate to NixOS by then.

https://github.com/nix-community/home-manager/issues/3968 - this to monitor and poke, there are also a link for a sample wrapper that might or night not help you.

You need this https://github.com/guibou/nixGL

Don’t use it - vote with your feet :)

Not quite. Suppose instead of a single version of serde there’s now 46 versions like in https://crates.io/crates/parquet to be able to use instances derived in some other crate X you have to use the same version of serde. Now, how should a crate decide which versions of serde to support?

All 46 and all optional? Supporting that would be painful. Just the last one? crates.io is a cemetery full of dead crates, with this support strategy any handful of crates picked at random are not going to be serde-compatible with each other.

A better solution would be a better support for compile time reflection so serde doesn’t have to exist in its current state, but that’s got delayed (by big macro conspiracy :)

urxvt. It works good enough and doesn’t use much memory.

deleted by creator

Kind of like flatpacks but it’s done with symlinks and fancy changes to the build systems. I think it fits better for the developer environment.