Hacker News Reader

some notes: - uptime kuma scales very bad, even with just 20 entries - Docker/Kubernetes is a trap, use something more suitable for a home lab like NixOS

SuperSandro2000on Music labels will regret coming for the Internet Archive, sound historian says

The greet of the music labels. They would likely not have made a penny of this because you can't even buy this stuff anymore!

SuperSandro2000on Trump's 'Crypto Reserve' Is Such Brazen Corruption

Who tells you that no one is ever going to change the code and just print more? It is just trust that this is not going to happen.

SuperSandro2000on The owner of ip4.me/ip6.me, Kevin Loch, has died

5 points on that some scammer is going to buy the domains and doing shady things with it.

SuperSandro2000on Federal workers ordered to return to offices without desks, Wi-Fi and lights

Doing great work at dismantling the US and making it a 3rd world country.

SuperSandro2000on Nix – Death by a Thousand Cuts

*it won't be in the future because it is no longer the grabbag for everything.

Also linking things to /usr/bin is done by the fhs which uses bubblewrap, not steam-run.

SuperSandro2000on Nix – Death by a Thousand Cuts

Be the change you want to see and help documenting the missing parts.

Also that is the old wiki but the same stays in the new one.

SuperSandro2000on Nix – Death by a Thousand Cuts

Everyone has some opinion. There are the people that say that flakes do to much and others are saying they do to little. No matter in which direction you go, someone is always unhappy.

Eval caching depends on pure eval and pure eval was previously not possible because channels are by design the most impure and cursed thing and are just a bandaid that lasted way to long.

The scaling with big repos gets worse as some use flakes instead of their normal build system which is not the intended use case. You still do development like normal with the normal build system.

Flakes are still unstable because everyone wants something different and there are still breaking changes planned which then everyone would wine about and maintaining backwards compatibility is a pain if you still change fundamental things.

SuperSandro2000on Nix – Death by a Thousand Cuts

Modern modules often have a settings option which directly links to the upstream documentation and you basically write the options from upstream into the settings option.

I look all the time in all kind of sources and skip all the tedious and annoying installation steps and just look at the described configs.

SuperSandro2000on Nix – Death by a Thousand Cuts

If you are using more than 2 channels than flakes are the natural progression.

SuperSandro2000on Nix – Death by a Thousand Cuts

When installing Nextcloud I basically have the following 4 options: - Do everything by hand and read through the docs on every update. Does not sound like fun. - Use someones Ansible playbook and hope that they update it on time. meh, also customizing it is not a walk in the park and requires some effort on my side. - Use the upstream Docker container which has the same customization problem as Ansible - Use the NixOS module. Updates are fast. Configuration changes are being handled by NixOS and I can easily inject a nginx location block in my declarative config. I also can easily describe extra bits like pre-compressed assets which then are served by nginx in my normal workflow without having to think about them at all on updates.

overlays and the module system are THE killer features. Almost no one else has something comparable to offer and if those powerful features are well understood, they can save you soo much hassle.

SuperSandro2000on Nix – Death by a Thousand Cuts

Yep, it feels somewhat clunky when you are used to NixOS modules :P