@skarn@discuss.tchncs.de @Solumbran@lemmy.world @linux@lemmy.ml

Have you considered the possibility that, by “finding a bug” and possibly “suggesting” a “patch”, the LLM could be smuggling another bug unbeknownst to the vibe coder(s) and/or smuggling a technical debt?

I say this as someone who’ve been coding since my 8s (now I’m 30), someone who hasn’t the tribalistic anti-AI sentiment (I even use LLMs sometimes, particularly the non-Western ones such as Deepseek and Qwen) but understands LLMs enough to know how the (current, state-of-the-art) stochastic parrots shouldn’t be trusted the source code of any slightly serious project, especially a full browser that Firefox is. Chances are devs are going to blindly trust and obediently stage-and-commit whatever the parroting machine spits out, and this can end up really messy. Given the ongoing pivot to AI from Mozilla, I doubt they’re worried about the consequences of vibe coding, though.

@dream_weasel@sh.itjust.works TIL Philipp Mainländer, David Benatar and countless other competent philosophers were/are all teenagers doing teenager ramblings!

@ennof@feddit.org @LuminousLuddite@lemmy.world @technology@lemmy.world

As if it were a matter of caring or wanting/not wanting to use websites… It would be really nice to live in such a world where one could have the luxury of “choosing”. Unfortunately, it’s not this world for many people and many peoples.

To exemplify this, there are websites I, as a Brazilian, can’t simply choose whether to use or not, because there are government and bureau websites for services through which I’m expected to comply with citizen things I didn’t ask for (as I didn’t ask to be born in this world to begin with). Online services such as “DETRAN” (state-wise transportation bureaus where one must renew one’s driver’s license), which I remember having to click a reCAPTCHA in order to proceed with transportation-related citizen duties. I can’t have the luxury of saying “you know what, I’m not renewing my driver’s license which has become my ID for a plethora of services not even related to driving, which means I’m going ID-less and becoming a legally-indigent person in the eyes of the next cop that requests my ID”.

Hell, I can’t even choose to have a degoogled phone because our customs (Receita Federal) will likely deny the entry for any “unlicensed device” (i.e. devices not licensed by ANATEL, Brazilian telecommunication agency). And installing a custom ROM in any available device is not without the risk of bricking the device (and losing a monthly minimum wage worth of money spent with said device) especially for someone like me who never installed custom ROMs.

Again, would be really awesome to live in the world you described where one could afford “caring to use” things…

@Grimy@lemmy.world @a_gee_dizzle@lemmy.ca

“Some free advice? You ever get back there, you hoard toilet paper. You understand me? Hoard it. Hoard it like it’s made of gold. 'Cause it is.” (Chuck Shurley, Supernatural TV series, S05E04)

@Solumbran@lemmy.world @pglpm@lemmy.ca @linux@programming.dev

The Brazilian flavor of age checking explicitly prohibits self-declaring (“vedada a autodeclaração”). Estimation of age via selfie or behavioral analysis, as well as the need for government-issued IDs, perhaps validation via credit card microtransactions, are some of the accepted age verification mechanisms for Lei 15211 (“ECA Digital” or, more informally known as “Lei Felca” due to the involvement of a YouTuber sub-celebrity on getting this thing to Brazilian lawmakers). Doing age bracketing via self-declared mechanisms, such as birthdate input or the usual consent button, risks fines and other provisions.

KYC (“Know Your Customer”) is, deep down, what these laws are going to be about, ID checks as sine qua non part of purposefully vague-worded laws with broad and outreaching enforcement, so tech organizations and companies worldwide, especially the smaller ones, will eventually find themselves in a situation where they are legally compelled to implement everything that’s being pushed as part of these dystopian laws. After all, it’s far from being just a Brazilian or a Californian thing.

Currently, yes, we’re seeing this law-concept restricted to a handful of places such as some USian states, as well as countries such as UK, Australia, Canada, now Brazil… Zoom out, however, and you’ll realize how this thing is gradually spreading worldwide because this is the only way for age verification to get effectively enforced.

You read it correctly, those laws are very likely getting to more and more countries, eventually turning KYC into part of international, industrial standards. Nothing too hard for big corps to do on their own, such as Google and Microsoft, even Canonical and Red Hat which are large companies, but small companies will end up being pushed into relying on non-free third-party KYC services in order to comply with age verification.

Such situation would end up benefiting the big players, with KYC services such as Persona becoming the new ubiquitous Cloudflare when it comes to this digital landscape. KYC gates, in this sense, would become the new CAPTCHA, Biometrics-as-a-service would become the new normal, true FOSS projects would become unlawful a priori while large corporations would thrive with another data point for tracking and advertising, and as the tolerance bar gets lowered, people will end up used to it, because any attempt to be against it will lead, at best, to social ostracization…

I don’t know, maybe I’m being overly pessimistic about it, but I can’t help but notice how dystopian things, some of which were long foretold and were warned about, are slowly taking away our privacy and freedom…

@LadyButterfly@reddthat.com @nostupidquestions@lemmy.world

There’s an eccentric hypothesis I thought of: maybe people used the fossils alongside the surrounding stones for buildings, without ever noticing the fossil. This makes me wonder how many ancient constructions, from simple huts all the way to entire castles and fortresses, contain fossils as stones.

And this doesn’t even seem to be limited to fossils: if we jump to Neolithic onwards, then fast-forward all the way to contemporaneity, some of the artifacts from back then (e.g. figurines such as Venus figurines, clay tablets, vases, papyri, petroglyphs, etc) likely ended up as part of buildings. Maybe those artifacts ended up unwillingly torn apart and ground by heavy machinery (e.g. backhoes, other earthmoving machinery, mining machinery and drilling machinery for petroleum wells, although these often involves prospecting, etc).

The artifact doesn’t even need to be that old: I once saw a news story about someone who used a “hammer” for decades before discovering it was actually a WWII grenade.

@1dalm@lemmy.today @linux@lemmy.ml

The country I exist in (Brazil) passed an age verification law “Lei 15.211/2025” which, to a certain extent, is even more dystopian than Californian one. Because, at least, the Californian “allows” self-declared age, while Brazilian don’t. This means systems must employ mechanisms such as ID’ing, age estimation by selfie or behavioral analysis.

When UK passed their law, threatening and, to certain extent, effectively sanctioning even even non-UK “disobedient”, something happened: many sites and platforms started to geoblock UK. Many Fediverse instances geoblocked UK.

Brazil has a similar history of legal outreach, we had court decisions trying to enforce and rule over non-Brazilians. Something similar is expected to happen when it comes to this age verification law. So I’d expect a similar widespread reaction of sites and platforms geoblocking Brazil.

In fact, it’s already happening: in mere two days since the law became effective, MidnightBSD geoblocked Brazil, Arch Linux 32-bits (not the mainstream Arch Linux) geoblocked Brazil, and others are expected to follow, both distros and websites as well. Including the Fediverse.

This kind of law will hardly stay in the countries and USian states where they’ve been implemented. It’ll spread, because the narrative it’s wrapped with is too alluring and compelling (from emotional appealing “Think about the children!?” all the way to the strawman “If you disagree with age checking laws, you’re literally a pdf file”). So expect more countries embracing this dystopia. This means fewer and fewer places where it’s not a thing. It reeks of a coordinated agenda, especially because it achieves similar things that intended by projects such as Chat Control, PIPA/SOPA, among many other previous authoritarian attempts. The authoritarian found the correct recipe: wrap 1984 in a cute “children protection” wrapping, rinse and repeat.

Therefore, some Fediverse instances, especially those sitting under the hurricane’s eye (e.g. Lemmy Brasil) may end up implementing age checking, or stopping altogether if they can’t afford the additional costs of age checking (it won’t be a free thing for platforms to do; a trivial cost for giants such as Meta, Google and Microsoft, but unfeasible for, say, Fediverse instances and FOSS projects).

Now, regarding the “kid friendly” limitation: if the Web gets limited to “non-adult content”… what’s “adult content” to begin with? Is it just porn, or it may end up covering several non-pornographic things?

It turns out, and here I’m risking getting too off-topic, many things would end up beneath this purposefully vague terminology “adult content”, content from many vulnerable groups: LGBTQIA+ (check out what happened during the recent itch.io and Steam crusade against “adult games”), women, pagans/occultists, political dissidents and whistleblowers, among others. This is what age verification laws are about: silencing everything deemed non-normative.

@Alaknar@sopuli.xyz @programmer_humor@programming.dev

Don’t take this as an attack

It’s okay, no offense taken.

where are you getting your news from?

Mostly from Lemmy, but also from Gizmodo.

Do people seriously believe that everything is AI in Windows now?

Tbf, it doesn’t help the fact that corps are shoving AI into everything they can and can’t. I’m far from being Anti-AI, but when we live in a world where everything is being AI-fied, I can totally understand the anti-AI fellows and their sentiment “Windows = AI”.

Recall is not yet live (it’s available as a preview feature), you have to enable it manually, and even then you can disable it easily.

As far as I read, it’s partially true. Not true, however, in cases when the PC was set up by someone other than you, e.g. in workplaces. If the company someone works to decides to enable Recall “to improve productivity”, anything done by the employee will be seen, not just by the employer, but by Microsoft too, not to mention hackers who will love to get their hands at this golden goose of private data.

They’re basically shoving the button wherever they can to goad people into using it, but that’s mostly it
It’s a button. […] Unless you click it, it does nothing other than taking up space.

Maybe. But the presence of the button, alongside the shortcuts for features “summarizing”, “auto-formatting text” and other AI-driven features, implies Copilot is a whole dependency on .dll/.exe related to Copilot, as well as potential unintended network comm with Microsoft servers.

“Purging” the OS from “AI-related crap” would purge it from AI-related crap and not break anything (source: did this on my previous work laptop)

Okay, fair point.

I agree about all the opt-out/opt-in stuff, but also understand why a company catering to 80% of the market defaults to opt-out - users are dumb, they have no clue how to explore features, so opt-in features remain forever disabled for 99% of them.

I heard the same during a discussion about Firefox here in Lemmy. “Users are dumb, so corp needs to guide them through the new features by enrolling them automatically”. Whenever I hear about how “users are dumb”, I can’t help but wonder how the so-called “dumb users” are allowed and able to drive a half-tonne car at 120kph or, even worse, (it doesn’t even need a license) voting (allowed responsibility over everyone’s lives)!

And then Apple does an update with an identical feature enabled by default, and everybody goes “damn, if only Microsoft thought of this!”

Maybe it’s because iThings aren’t socially pushed as Microsoft things are. You said yourself: Microsoft is “a company catering to 80% of the market” dominating the PC market, not Apple.

what “crime” you mean

Non-consensual relationship. Harassment. In this case, it’s software harassment disguised as dark patterns such as opt-off when it should be opt-in.

@Alaknar@sopuli.xyz @programmer_humor@programming.dev

Well, I confess I’m in no good position to say anything about Windows 11, for I’ve been a daily Linux-only user (Arch, by the way) for almost a decade.

However, as far as I’ve seen about Windows, AI (especially that spying feature designed to take screenshots and create a lookup-able timeline, “Microsoft Recall” if I recall (pun intended) correctly) seems to be so intertwined with Windows that even the Windows Explorer’s Shell has now a hard dependency on AI-related and Microsoft Edge-related libraries. This way, if someone were to try and purge the Windows from AI-related crap, it will break the OS, Explorer simply won’t launch.

Also, “can be easily turned off” implies something that comes enabled by default: the exact same dilemma behind Mozilla Firefox and all the crap they’ve been imbuing inside the browser. In the end of the day, it’s a non-consented relation. The fact it can be opted-out doesn’t make it less of a non-consented relationship, for the non-consented relation already happened as the user proceeds to opting-out of it. In other realms of legality, it would be considered a crime, but as it’s something done by corporations (Microsoft, Mozilla, Google), they it’s suddenly “a-okay”.

@flambonkscious@sh.itjust.works @programmer_humor@programming.dev

Oh, I misremembered the file extension. Yeah, that’s right, it was edit.com, there was likely no .exe because it was a relic from before Windows NT.

And, BASIC… Such great times. Although the BASIC flavor I dealt the most with was Visual Basic (VB5 and VB6), I also did some tinkering with terminal-based BASIC flavors (specifically, Linux ports of BASIC interpreters) as well.

@rabiezaater@piefed.social @nostupidquestions@lemmy.world

generating ideas

LLMs don’t generate ideas, stricto sensu. They do, and I find it useful for esoteric (gnosis through chaos magick) purposes, output names and words unbeknownst to the user (this is how I, as an ESL person, learned some words I didn’t know before).

But if we consider hard determinism, do we as biological automatons, though?

learn to code

As someone who codes since my childhood, I wouldn’t suggest relying on LLMs for that. They could be used to output a descriptive text about some function or library, but you must know LLMs are statistical machines, the output text is a chain of “which token is the most probable next?”, an auto-completing only slightly “better” than, say, Gboard’s auto-complete. They “hallucinate” precisely because they rely on statistics and randomness.

Again: extremely useful as an “Ouija board”, not very useful for blindly relying for learning something, definitely not reliable for “vibe coding”.

Wanna learn how to code? Do the Elliot Alderson (Mr. Robot TV series) approach: find an existing “Hello world” project/source-code, tinker with it, change things here and there, try to compile/run, Google the exception that the compiler/interpreter thrown at you, change more things, break things, then fix the things you broke… This is exactly how I did. Let go of any hurry and you’ll likely going to master it eventually.

d&d […] I need a character […] it makes it up quick

Yes, this is one of the use cases where LLMs can thrive, as a dice with hundreds of billions of sides.

You may want to roll real dices, convert the number into the respective letter (A=1,B=2,…) then append it as a source of real entropy, because the randomness you get from LLMs is likely to be pseudorandom.

Ideally, you’d tune (using a RTL-SDR) to a blank radio frequency and digitize the (true noise) spectrum into ASCII, and voila: free randomness, straight from the Cosmic Womb to your computer!

get upset about AI “stealing” work with regard to code or other stuff that people willingly put out there for free for others to consume

Totally agree with you in this regard. Throughout the history, humans relied on other humans’ “ideas”. Most of the novelty stemmed from “what if I were to take this flamey thing that consumed the tree I used to sit on, and put it under this food?”, mashing up existing things. If we really were to appeal, evolution is that, merging two genetic sequences in an approximate manner while trying to replicate, still I don’t see humans accusing newborn of “stealing genetic work from their ancestors”.

definitely useful in a lot of ways, […] if […] developed on a more localized and decentralized scale

I totally agree in this regard, too.

To answer the main question: IMHO, people hate AI because it has been pushed and used by corps to further enshittify this world. I’m not Anti-AI, but I’m not pro-AI either. There can be nuance from both.

@AnUnusualRelic@lemmy.world @programmer_humor@programming.dev

Good question. I’m not sure. I guess no, because, as far as I know, ed is a GNU editor which allows for composing and editing files in a REPL-like environment (whose specific commands, apart from “q” to quit, I’m yet to learn)

The “edit” I’m referring to was a spiritual antecessor or cousin of vim, emacs and nano. It was a TUI, full with a functional menubar accessible through keyboard arrow keys. I remember it having a blue background with gray/white text.

I remember with quite a certainty it was a thing for Windows XP. Was invokeable by using “edit filename.txt” in cmd.

However, I also remember having manually copied some executables across diferent Windows versions in order to test and see whether these old executables would work. I remember having successfully ran Windows XP’s calc.exe in some later Windows version, relying on the compatibility layer (“ntvdm”, I guess?). I remember doing the same for 16-bit, MS-DOS programs, but I don’t remember whether “edit” MS-DOS programs was included in post-XP Windows versions, or if I manually copied it from XP.

@RmDebArc_5@piefed.zip @programmer_humor@programming.dev

Wasn’t there a lurking edit.exe or edit.cmd somewhere inside C:\WINDOWS\system32? Would make an interesting replacement to the enshittified “Not-e-pad”. But, then, I haven’t used Windows since Windows 10 was still a novelty (and what definitely pushed me to Linux… Arch Linux btw), so maybe I’m very old (“I’m old, Dean, very old”) to recall of a MS-DOS relic.

@REDACTED@infosec.pub @RmDebArc_5@piefed.zip
@programmer_humor@programming.dev

I’d say LLM is a notepad client now.

@ominouslemon@sh.itjust.works @linux@lemmy.ml

The git PR specifically mentions a birthDate, a data struct that feels like it could easily be tampered with (therefore, far from “confiável” (trustworthy) as explicitly required by “deverão ser adotados mecanismos confiáveis de verificação de idade” (“trustworthy age checking mechanisms must be adopted”)).

Thinking of age checking as some kind of OAuth flow, one would ideally store the authz token from whatever age checking provider validated the user’s age, instead of some plain data which, depending on the provider, wouldn’t even be handed to the application.

I can sort of imagine the following, hypothetical flow:

1. Human tries to access the system for the first time
2. System asks for human consent to proceed with age checking
3. Human (is compelled to) accept going through age checking shenanigans
4. System redirects human to 3rd-party age checking provider interface (e.g. Persona).
5. Provider proceeds with whatever means necessary for the human to upload ID and/or selfie, who does whatever is required from them by the provider interface.
6. In case of IDs, the provider talks with gov databases (e.g. Receita Federal do Brasil for CPF “Cadastro de Pessoa Física”) in order to attest the validity of the ID. In case of selfie, provider communicates with a facial recognition model/algorithm/platform.
7. Provider gets the information necessary for age-bracketing, appends it to their own DB with a signing hash, then returns the digest of said hash as a token to the system.
8. System receives the authorization payload and confirms with the provider whether it’s a valid token.
9. Provider replies positively, perhaps with some kind of checksum, regarding validity of the token.
10. System stores the token to hand it to whatever subsystem (for OSes, a software; for online platforms such as social media, a module/route) requesting age info.
11. Subsystem allows or denies human access.

Some age checking models (such as EU) seems to be doing a similar thing to what I hypothesized above: the EU Digital Wallet returns a token, instead of PII. A token that can be checked against the Digital Wallet API for validity (theoretically) without disclosing who the user is (in practice, it’d be another, pretty reliable piece of traceable data despite any “anonymity”)

I’m not sure whether a similar thing will be implemented here in Brazil (we got an official gov app, gov.br, which can already be used for “social log-in” by 3rd-party platforms, but I don’t know whether it’s ready for age check provisioning).

As far as I know Brazil and Brazilians, it’s highly likely we’d end up with dependencies on Microsoft or Google services because Brazilian gov can’t help but handing its own sovereignty to US tech corps, which adds to the dystopia.

I must make something very clear: I’m far from agreeing with this dystopia, I deeply despise this whole “age check” thing going on worldwide; I’m just thinking as a DevOps would.

@skyline2@lemmy.dbzer0.com @linux@lemmy.ml

Brazilian here. I’m neither a lawyer nor a specialist, just someone who has read the Portuguese text from the Brazilian flavor of the ongoing worldwide age check set of laws.

I must note that the Brazilian age check law (Lei 15.211/2025) specifies “vedada a autodeclaração” (English: “self-declaring is forbidden”). This means that this kind of implementation, where age or birthday is an user input, wouldn’t be compliant to Lei 15.211/2025, because it requires the age information to be assessed independently from the user whose age is being assessed. This means face biometrics, government-issued ID (in our case, CPF, CNH, Passaporte and similar) or “behaviorial analysis”… Anything but a “yes I’m 18” or “I was born in day month year”, for those are self-declared and the Law says it’s “not enough”.

Someone should warn the systemd maintainers of this “Brazilian jabuticaba”.

(Cross posting this reply of mine because the post was cross posted to two different Lemmy instances)

@DeathByBigSad@sh.itjust.works @nostupidquestions@lemmy.world

Sounds like emotional blackmailing.

@oopsgodisdeadmybad@lemmy.zip @Krauerking@lemy.lol
@aboringdystopia@lemmy.world

First, I don’t smoke and never had smoked (not considering the passive smoking from being close to smokers), the only “drug” I use non-frequently is alcohol (a vodka-saké beverage).

This said, your paternalistic implied thought (“drugs are bad, people shouldn’t use drugs”; are their bodies yours, to begin with? Do you happen to have some kind of ownership over other’s bodies? Who gave you this almighty authority, do you happen to be a Demiurgal Archon of sorts?) is missing the forest for the trees.

I’m barely knowledgeable about biology, but I know enough to remember about symbiotic relationships in nature, as well as phylogenetics. The plant responsible for tobacco wasn’t some kind of lab creation, IIRC tobacco plant even predates us homininae, and this means there are certainly species biologically interacting with the tobacco plant, perhaps species reliant of tobacco plant (not necessarily because of nicotine, but because it’s a plant which siphons nutrients and resources from the soil, displacing water and somehow contributing to water cycles).

The moment this Archonic hubris illustrated by your text were to render tobacco extinct, unbeknownst to you, other species would be going extinct as well, and this may include humans due to how food web is so interconnected in ways we’re still trying to figure out. That’s part of why we try to classify species taxonomically and phylogenetically, because this is also valuable for us to identify potential interspecies relationships. I mean, this is exactly what phylogenetics is about: study of evolutionary relationships among biological entities.

So, please, seek some phylogenetics and biology awareness before feeding paternalistic, bigoted anti-drug thoughts. Others bodies aren’t yours for you to wish for control, let alone other species.

@Krauerking@lemy.lol @technology@lemmy.world

Polish Roman Catholic background. Not exactly tolerant either and now I live in a place that famously is Mennonite and Amish.

Oh… I’ve heard about the Amish… So it’s practically a similar situation over there.

If they will make you suffer why not suffer on your own terms for a cause you think worth it? Why wait to be starved by someone’s hand other than your own?

Yeah, you’re right. There’s a part of me who thinks this way, too. I guess this whole age check thing will inevitably push me in this regard.

I wish you safety and health though.

Thanks! Wish you the same.

@SalamenceFury@piefed.social @qwerty@discuss.tchncs.de
@technology@lemmy.world

The news articles about this law, if said articles were published, are likely buried under the ongoing Caso do Banco Master (a large financial scandal involving a bank), the all-encompassing political crisis going on in Brazil, the international Iran-USA conflict, among other ongoing events. There are too many things happening simultaneously, so I don’t really blame news outlets: they can only cover so much because we, as humans, can’t be aware of all things when too many things are happening. So this is why little (if anything) about said law is being reported by news outlets such as Globo/G1.

Even as a Brazilian myself, I wasn’t aware of this law (I was only aware of the so-called “Lei Felca” named after the YouTuber/TikToker Felca; but it doesn’t seem to be this law specifically). I only got to discover about this law through the English-speaking Fediverse and Nostr posts.