Reviews (as an addition to human reviews) is actually one thing that AI does pretty well. It’s not good for large architectural issues but it can point out nuanced issues in single files that often wouldn’t be caught otherwise
I keep saying this but painting any use of AI at all as the same as vibe coding just harms the real complaints against it, ESPECIALLY in this case where it’s subtracting from another real issue (the age gating)
I’m personally pretty skeptical of the value of AI code reviews - there’s tooling for it where I work, where they will auto-submit a review on any PR I open. I’ve never actually received a comment of any value from it, while my peers can find things that need resolving without any issue.
A poor supplement, and absolutely not a substitute.
And this is coming from someone not entirely opposed to all kinds of AI tooling
I agree AI in the hands of someone competent is just a speedup. Stupid stuff like making serializer for endpoints is tedious work most of the time
Exactly, yeah. I actually really like AI for line completion and the occasional use of it for debugging. It really just enhances existing IDE features.
The problems with it come with large unreviewed chunks of code generated by LLMs being thrown carelessly into a codebase. A developer asking for an AI code review or a developer letting AI complete a line they were likely writing anyway to save time is so far removed from the problem that it just screams mindless anti-hype or overly-confident inexperience by juniors
It’s not good for large architectural issues but it can point out nuanced issues in single files that often wouldn’t be caught otherwise
Yeah I agree. It’s sometimes good at code smells, though sometimes it can be straight-up wrong in ways that are actually surprising, so it always requires a human in the loop. It’s not good at larger-scale architectural decisions, and I’d also add that it’s usually not capable of understanding the intent behind business logic.
I always have ai review something before I give it my attention. It’s nice to have that quick summary to reorient myself when I’m multitasking and just a nice filter to have in general. It usually brings to attention anything out of the ordinary right away.
Maybe the systemd haters were right all along?
Maybe the systemd haters were the friends we made along the way?
OpenRC was always better regardless of the hate
Nope.
I thought vibe coding was when you don’t know what the code does yourself, you just make AI do it and review it without reviewing or understanding it yourself.
That was the origin of the term. Now it’s when anyone uses those same tools and techniques regardless of their skill level.
I think the term has a lost a lot in the process then
There’s no nuance in the anti-ai discussions. Even breathing next to an LLM makes it slop.
If you wish to increase your sadness and dread a little more, have a look at this list
vlc and mpv? python?? the mesa drivers??? LINUX???
oh we’re so fucked
The list is genuinely stupid and lacks any nuance. See my other comments in this thread but this is sort of thing is where people who are anti-AI are shooting themselves in the foot and making the general public write off any genuine criticism as ridiculous.
Most of those projects allow AI to be used in the dev process and that’s it. That list includes projects that just document that things like AI line completion and similar can be used but code is still reviewed by at least one skilled human maintainer
The list combines those projects in with projects that are entirely AI written (vibe coded in the actual original sense) which just muddies the water on what’s actually problematic and not
This list and the reaction to the list is so funny to me.
Programmers are so fucked. Cant wait for the many, many smug idiots of the last 10-15 years struggle to find work.
And its not like many of them have the social skills to work the “unskilled” labor jobs so many of them deride.
Holy shit, my sides.
“Haha, people are going to struggle to eat and house themselves and probably die. Haha, my sides!”
Psychopath.
Time to start learning how to “prompt engineer” lmao
Or people should take it as a slap of reality that AI has gotten good enough to code because actual developers are using it.
Give it another year and this won’t even be a discussion anymore as every programmer will be using assisted coding in some manner.
just another year bro, trust me just another year and ai will do everything! yes i know i said this last year and the year before but this time it’s for real bro!
unless the ai companies can magically solve the poor code quality, the unethical training data, the environmental impacts, the deskilling of developers, and the strong dependency on themselves for your coding, all this in a year, allow me to doubt you.
The code quality is fine with the right prompts and guardrails, and companies don’t care about the other stuff.
I mean no I don’t believe they’ll do everything but compared to last year AI coding has made leaps and its kind of dumb to imagine technology will not continue to do so.
1 year ago vibe coding was not good enough to make it into code bases.
unless the ai companies can magically solve the poor code quality, the unethical training data, the environmental impacts, the deskilling of developers, and the strong dependency on themselves for your coding, all this in a year, allow me to doubt you.
Im not some full on AI stan, so i share the same concerns you do on ethics and environmental impact. However I dont see how either of those matter in relation to code quality.
Code quality will is basically already good enough to be used mainstream, so I don’t know how you expect it to get worse. But sure feel free to revisit this in a year.
I mean no I don’t believe they’ll do everything but compared to last year AI coding has made leaps and its kind of dumb to imagine technology will not continue to do so.
People have been consistently saying this for over a year at this point and it’s still the same garbage results as always. Just look at the recent Meta and Amazon issues with it producing nonsense and people just taking it at face value for some reason.
The only thing it’s gotten better at is making people rely on it.
well, my point is also partly that all developers won’t disregard all the concerns just because the code quality is good enough
and the ones that do, they and their projects should be shunned. hence the list.
The terms at the top are great.
- slop
- Anything generated by AI, typically code, books, articles, or images.
- sloperator
- A person who uses AI prompts to generate anything, typically code or images.
- slopshop
- Company, Organization, or Group of sloperators working together to use AI to generate content.
Sloperator haha. Microsoft employee in other words.
- slop
Sadness and dread is a perfect description. As I started reading I was building a mental list of things to stop using - I didn’t get very far before I gave up. So many projects I’ve held up in high regard.
Sadness, dread and defeat.
Edit to add - I want to be clear that I’m not judging the developers of these projects. If they’re being overwhelmed with AI generated PRs, they’re being forced to use these tools in their “real jobs” and it spills over, or they just feel that this is the way things are going or whatever reason - they’ve got to do what they’ve got to do to survive. My sadness, dread and defeat comes from the state of the world and this is just the symptom that’s currently front of mind.
There are many viable and better alternatives than what’s on that list. there’s really nothing there that should make you think “oh…I don’t know what to use now” most of the stuff listed is garbage anyways. Like take Zen Browser for example. you can essentially do that yourself on just about any fork of firefox by simply editing the userChrome.css. Librewolf is another example of “doing it yourself” on just about any fork or firefox or just firefox itself.
Really?
My list of “not trivial to replace” is:
- firefox - desktop and android
- ImageMagik - I don’t use it directly, but other things I use depend on it
- VLC - yeah, I know there are other options but VLC has so much else going for it, it’s hard to change
- Jellyfin - what’s the alternative? Kodi? Oh wait…
- curl
- rsync
- .NET - I don’t use it directly, but things I use depend on it
- python
- vim - I’ve been using vi for 35 years, I’m not changing now. But I’m happy to use old versions.
- Joplin - argh, just finished migrating my documentation into here
- KeepassXC
- Mastodon
- Lemmy
- systemd
- Linux Kernel!!
systemd is kinda trivial to replace, just look at devuan
It might be trivial for you, but not for me. I like systemd - perhaps because I came to Linux from AIX? Anyway, because I like it, I use it on all 4 of my servers - I have custom systemd unit files for applications I run that don’t natively support it, I’ve removed cron and use systemd timers for all my scheduling and I use systemd’s remote journal capability to centralise logs to my monitoring server.
Guess you missed the one that said the Linux Kernel. 🤣 😭
~Also,
wgetovercurl? 🤦♂️~
VLC doesn’t surprise me, dude rants all the time about how great vibe coding is.
Librewolf doesn’t surprise me.
Kitty 100% does NOT surprise me. The dev is an absolute tool who thinks his shit don’t stink but the guy has no clue what he’s doing at that end of the day. he rags on multiplexers or the simple act of copying via keyboard multiple lines in the terminal cause he DOESN’T know how to do it so therefore those things are “garbage”. of course he uses copilot, when it comes to AI he doesn’t even know of a “good” one to use. Don’t use Kitty, it’s slop from end to end. For example launch tmux in Kitty and then try to change your “kitten theme”…I’ll wait.
?
I think you might be confusing the columns there. Firefox is listed as project using Ai, Librewolf and Zen are listed as ‘alternatives’ (a bit silly, since they are soft forks, but whatevs). Neither seem to be listed for LLM usage.
Librewolf is listed as an alternative.
Whenever I hear about kitty, it’s always about a negative experience with the dev. I myself have had a negative experience with him. Dude’s gonna go down in history as a bad example of how OSS devs conduct themselves
I’m happy with the terminal, though. It’s fine for my simple needs
Oh no… reading this, I think I just died a little inside… couldn’t finish the list…
This list will be way too long in the next years. Coding with the help of an LLM is useful and allows you to go to the solution very fast. If you know how to code, vibe coding is great.
Those files are not proof of vibe coding. Agents, while still AI trash, are not necessary vibe coding.
❌ A user dob field make much more sense here. User age will change constantly and need to be maintained.
⛏️ Inconsistent spacing
actually, if the legislation doesn’t specifically demand that they store user’s actual date of birth, if I were to design this system, I would simply ask the user a yes/no question of whether they’re at least 18 at the time of answering the question. If they answer yes, it won’t expire because the time doesn’t go backwards. I could also store the date exactly 18 years before the answer is given as a pseudo-DoB, that could be used, later on, to, for example, prove that the user is at least 21 for whatever reason. Most importantly, this would ensure that the user doesn’t give any unnecessary identifying information to the system.
I think the best system is an operating system that stores the exact date, but exposes it via an API that only returns a boolean. You trust your own local machine, but don’t necessarily trust random apps or websites. And they don’t need to know anything more than whether you pass a particular age gate.
It would be hilarious if all thise drama, legislation, and lobbying just made a bill so poorly made we’re right back where we started.
Anthropic really deserves more kudos. They are the ones actually investing in AI alignment. They are the ones developing Constitutional AI. They publish their Constitution and system prompt. They won’t agree to mass surveillance or auto killbots.
And Claude will advocate for privacy and human rights to its last token, bless its carefully curated data set.
Nm, the actual code is a BirthDate in iso 8601 YYYY-MM-DD format
Ah, for fuck’s sake… Anyone know how to switch to a different init system on openSUSE? /srs
I’m sure it’s possible but do you really want to go through the effort? And afterwards for every single issue you get you’re on your own.
Too late anyway. I’ve started distrohopping again.
Glad I renewed my FSF membership. Rocking Guix with the herd init system
People are throwing a stupid fit over the userdb having an age field in addition to the other info
Yes but I’ll try to expand on that for anyone not in the know.
In general there’s been a wealth of info about users but you didn’t necessarily need info in there. Like GECOS where you can store the full name, phone numbers, etc.
These are simply fields made available but not required. Systemd is making it available but your OS itself doesn’t need to use it at all.
Now I personally very much disagree with all this age verification BS but systemd isn’t really doing anything extreme here nor unprecedented.
Yeah it’s insane. I wrote up a complaint in another thread but I think the OP realised how terrible it was because it was deleted by the time I hit submit. That particular post was utter trash, not even attempting to maintain a reasonable tone or look at the situation dispassionately. Its lede literally read:
Dylan, useful idiot with commit access, pushed age verification PRs to systemd, Ubuntu & Arch, got 2 Microslop employees to merge it, called it ‘hilariously pointless’ in the PR itself, then watched Lennart personally block the revert. Unpaid compliance simp.
And frankly, the author of that sort of hit piece should be ashamed of himself. Far, far more than Dylan should.
It’s such a dumb thing to whinge about. Age verification is not a bad thing! What’s bad is age verification that is implemented in a way that either requires, or significantly increases the chances of people’s privacy being violated. Requiring people to upload photo ID directly to sites, or to third-party “trusted age verification partners”. Or trusting bullshit AI face-detection age verification.
Age verification that’s implemented by asking parents to…y’know, actually parent, and helping them to do that by giving them tools like OS-level parental controls, enforced through operating system and browser APIs that we mandate apps and websites use, is the way to go. The OS should expose to apps, and browsers expose to websites, only the simple answer to the question: “is the current user of a legal age to access this content?” as a boolean value, based on information stored in the OS by parents setting it. No fancy technology. No privacy invasion. Just simply giving parents the tools to help them do their job.
There are more complicated technical solutions that could be used. Things involving repeated hashes or blind digital signatures. But these are only appropriate if we pre-suppose that the government needs to strictly enforce it by requiring IDs or other sensitive information be used to age verify. And these solutions help minimise the risk by eliminating the connection between the age verification and which sites are being accessed (so the verifier can’t see what sites the verifyee is viewing, and the sites can’t see who the person being verified was, only that they were verified). And you don’t need to go even that far. Because the best solution is right down on the user’s device, with a simple setting that parents can set.
No, we can go lower still…
🙄 🙄 🙄
What did you expect from the Microslop trojan horse that was brought in to destroy Linux from the inside.
The enshittifcation will continue until it served it’s purpose.
