I do have formal education in the field, but that doesn’t mean that I’m easily capable of e.g. writing a module for Nginx or even most website frameworks or knowledgable enough about administration to feel comfortable exposing a dynamic website (as opposed to static HTML+CSS) to the internet.

A lot of it is learning by doing, though, and for a lot of things books are a great resource (some stuff is outdated as soon as it’s published, but general system administration or programming knowledge doesn’t get outdated that fast).

Didn’t know that one before, I didn’t even consider the dependency issue!

But, yes. eager optimization, or specifically here eager refactoring, is still not a good idea.

Beware premature optimization, indeed! Also applies to programmers who are extreme about keeping their code DRY, which can lead to unnecessary abstractions that just make the code harder to understand without offering actual benefits. IME some languages suffer from this more than others, e.g. Ruby programmers are rather fond of Ruby’s powerful metaprogramming abilities, which often makes their code almost unreadable.

Sounds neat. KDE is a pretty important part of the Linux desktop ecosystem and lots of people are running fixed-release distros.

That’s like yesterday by Debian standards, lol. And honestly, I’m surprised that they’re adding this kind of feature after this long. Feels similar to how i3-wm eventually merged gaps functionality (as known from the popular i3-gaps patch by airblader), I never thought it would actually happen.

LTS Debian

Why would you run any other kind of Debian? I’m proudly running oldstable, my laptop has an Ubuntu version that’s even older. And though tempting, I’m not upgrading for nano -/.

I really appreciate the info, though!

Well, micro has syntax highlighting and LSP support, and vim is just really efficient at text editing … though nano’s feature set is honestly fine for what it is, I’d much rather it used well-known shortcuts by default.

sudo apt install micro

, micro is nearly seven times the size of nano,

Still starts up instantly, though, and still smaller than practically every GUI editor.

-/

$ nano -/ test.txt 
nano: invalid option -- '/'

So much for that. Is this a really new option, or does it require a compilation flag that Debian doesn’t bother with?

The UI isn’t the issue with nano, it’s that it needs that kind of UI … you could memorize nano’s shortcuts, but it doesn’t have enough features to justify the effort.

Big fan of micro. Like nano, it’s non-modal, but unlike nano, it doesn’t need to print its shortcuts on the bottom because everyone already knows most of the shortcuts from almost every GUI editor. And it has inbuilt syntax highlighting and LSP support (TBH it’s a bit much at times, the UI is a bit too intrusive to (by default) complain every time you violate some style guideline).

Though that’s moot if you don’t have admin privileges on the box you’re working on.

A language’s packages need to be in lockstep with the language’s version, and they usually depend on each other, too (to a much higher degree than they depend on programs/libraries from other languages, not counting stuff like C that many higher-level languages are built on).

Also gonna get really bad once these lonely maintainers get into AI. Been browsing https://codeberg.org/ethical-foss/open-slopware, things looks pretty bleak.

Pretty sure lots of projects use json files, before that it was probably XML. Are those good or even decent alternatives? Probably depends on the amount of data …

And what most websites doing is actually not allowed according to the GDPR, it’s supposed to be a one-click opt-out (which would be easy to automate with a browser extension). Unfortunately, the relevant court is in Ireland, which is the EU’s hub for US big tech.

Now explain why tankies seem to be extremely fond of current Russia, even though it makes no attempt to hide its rightwing authoritarian nature. “Geopolitics”/“enemy of my enemy” can only explain so much of that IMO.

Yeah, keeping the needed packages around would ballon storage size if you want to do it for an entire directory of distro isos, especially if you’re interested in historic releases (e.g. for hardware compatibility). I’ve actually been experimenting with using locally saved packages lately, mostly because HDD+ethernet is about 20 times faster than downloading packages off the internet for me. It’s fine as a cache if you’re managing several instances of the same distro, but if you want to full-on install several different distros/releases off HDD you’ll need to seriously invest in storage space. Might be worth it though … hmmmmm.

Impressive effort.

Somehow, I perfectly remember ncmpcpp (what does any of that even mean? something with C++ I guess?) but struggle with lighttpd (light+httd+daemon).

So … how enshittified is Apple nowadays? Last time I had to work on a Windows machine, I installed a Linux desktop in WSL, but those were the pre-LLM days.

Why is the article talking so much about the lack of new release tag if the rewrite is already being shipped?