Mikina

I don't think you need any active sabotaging in this regard. I'm not really worried about the future of LLMs, because we are already at a point of feedback cascade where thanks to LLMs, more and more of content they steal from the internet has been AI generated by them anyway, which will eventually cause the models to collapse or stagnate. And besides, you wouldn't be able to sabotage at a scale required for this. Thankfully, the spread of fake AI generated websites and content it has enabled is so massive, that it works as well.

I'm looking forward to that.

This. The whole discussion about "tinkering with immutable distros" fells like it misses the point and literal meaning of atomic and immutable.

Rebuilding the whole OS to layer another immutable read-only part into it isn't tinkering. Of changing one OS file has you rebooting, then that's not tinker-friendly.

Atomic distributions are by definition something you don't tinker with, and it stays the way you need it.

And no, having bundled distrobox or rollbacks doesn't make it tinker friendly, you can do both on normal distribution.

But once you have done tinkering and want the system to stay the way it is - that's what atomic means and is for.

Snapshots and rollbacks already exist in other distros, so the (only?) advantage you are mentioning is kind of a weak point.

Everything is a bother, since you can't just easily dnf install what you need, without actually rebooting or dealing with containers. I wasn't able to get a Win11 VM and work VPN properly working for long enough that rebooting to windows and just doing the RDP there was easier for me. (Because getting TPM to work simply wasn't feasible on atomic, and no - FOSS rdps didn't work)

If an app doesn't have Snap or .App file, it will be a bother. Having to enter a container just so I can edit something in a properly set up nvim just sucks, adding bloat to something that could have been one easy command.

There's a learning curve that gets in your way a lot, and since there are no actually payoffs for going through it, why bother?

I currently have Bazzite on my desktop as a daily driver, and it has been way worse experience than I had with Nobara, debugging any issues with I.e audio or drivers is awfull because the resources about it are a lot sparser, and so far I simply don't see anything it does better. I did rollback my Nobara few times with brtfs and it never was an issue.

One thing that may be worth it, if it's the case - can you actually export your layers into a VCS that you can then simply clone, just like you can with NixOS? Because if not, then following your logic, there's really no point in choosing atomic distro over NixOS. Sure, it has a slight learning curve, but you get a system you can not only rollback, but also easily clone anywhere you need it. What are your reasons for not using NixOS?

That said - there is one use case where atomic distros are amazing - if you have a, well, atomic environment you don't need to change often. Bazzite on SteamDeck or LegionGo being the best example, I'm using it there and it's been amazing experience.

This... Is actually unironically the best argument I've heard in favor of AIs so far, that I haven't thought about.

Still - the thing you'd be doing instead is feeding money and attentention to AI bros, and that's probably even worse than any job you could be micro sabotaging.

How's PostmarketOS doing recently, anyone using it as a daily driver? I have a PinePhone in a cupboard that I bought more than a year ago that lasted like a week as my daily before I quickly gave up on it (or rather, reinstalled it to Kali Nethunter and just have it in my pentesting bag. Not that I ever used it :D), since it had too many issues.

My issue with canvas fingerprinting and, well, any other fingerprinting is that it makes the situation even worse. It plays right into the hands of data brokers, and is something I've been heavily fighting against, and simply don't visit any website that doesn't work in my browser that's trying hard not to be fingerprintable.

Just now there is an article on the front page of programming.net about how are data brokers boasting to have extreme amounts of data on almost every user of the internet. If the defense against bot will be based on fingerprinting, it will heavily discourage use of anti-fingerprinting methods, which in turn makes them way less effective - if you're one of the few people who isn't fingerprintable, then it doesn't matter that you have no fingeprint, because it makes it a fingerprint in itself.

So, please no. Eat away on my CPU however you want, but don't help the data brokers.

I did Software Engineering Bachelors and then gamedev masters, and while I didn't really appreciate it at the start, since it felt like I'm learning a lot of stuff I'd never need, I've eventually come to be really glad that I did it.

Throughout the classes it felt pretty meh, I didn't understand why I have to do so much stuff that I'll never really use, and always felt like I'm just forgetting 90% of what I was taught the moment I was done with finals for that class. Why do I need to learn Smalltalk? Why Lisp? What even is Prolog? Does anyone even do UML anymore? I want to be a C# programmer, I don't need this.

And it was true. From most of the languages I've had to go through, I don't remember almost anything. But that's not what it was about, and that's something I only came to appreciate with time - it was not about learning Pharo or Prolog, it was about overcoming the initial learning curve and getting somewhat familiar with OOP or formal-logic style of languages. And while you forget the details, the familiarity will stay with you. The goal is not to make you a Prolog programmer, but to make you a programmer.

I've eventually realized that I can pick up any language pretty quickly, no matter what it is - because I've already seen and learned all of the different styles or types of languages there are, and no matter what it is, it's similar to something I vaguely remember seeing somewhere. And that's an immense help. I picked that up naturally, I've kept hearing the question "what programmer are you? What language you can program in?", and it felt weird - sure, I do know the most about C#, but I never had issues with picking up whatever was close at hand or needed, and writing anything I needed with a little bit of documentation and googling. And it was thanks to what I learned in school.

And the same applies to the math and data structures that they hammer into you. Do I remember the difference between red and black tree, or a min-heap, and can I prove it? Not really, but I know they exist, and when I see a problem that sounds like it could use some obscure data-structure, it comes to my mind and I know what keywords to look up. And that's a skill that I've notice is missing from most of the people who didn't have formal CS background. Same goes for algorithms like FFT - you know it exists and what it's used for, and seeing a problem that could use it will trigger your PTSD.

So, I highly recommend giving college a try. You will learn a lot of cool algorithm, and some of the classes were fascinating, and it will give you a vague overview that will stay with you throughout your carreer, feeding you with keywords about stuff that might be usefull for the problem at hand. It's the best thing I've done in regards to programming.

This is the worst way how to announce something like this.

I don't know the context, but if the goal was to not start a wave of speculations, it would be better to simply not hint at anything. I wonder what happened, and I respect if they don't want to deal with it, but this does feel weird.

Is this reply AI generated?

What would be ELI5 use case of this? It has been almost a decade since I did anything math-formal in college, and I wonder what would be some practical uses or situations is SW dev where you should turn to this language.

EDIT: I skimmed the intro to Verifiable C, and I think I vaguely understand the idea - do I get it right, that the point is to basically create a formal definition of the function you are writing, i.e if you have a function that takes an array and sorts it, you'd have something like

For every sequence a and every i, 0 <= i < len(F(a)) -> F(a)~i~ < F(a)~i+1~

(Or whatever would the correct formal definition be, I don't really remember the details, I know I missed some stuff about properly defining the variables, but the idea of the definition should be kind of correct)

And then you define this formal definiton in CoQ, then somehow convert your code into CoQ code it can accept it as F(a), and CoQ will try to proove formally that the function behavior is correct?

So, it's basically more robust Unit Testing that's backed by formal math proofs?

I think I know who killed him.

By 11, he was programming on his own—a skill he used to playfully torment his friends. One remembers Balaji’s idea of a middle-school prank: writing code that deleted a friend’s Skyrim save file.

but if they do it’s a scandal waiting to happen

That was my line of thought. If you pay for failed captchas, there are a few websites using it that'd deserve a bot failing them constantly.