A year or two ago I noted that LLM answers to general queries seemed to be wrong 1/3 of the time, right 1/3 of the time so far as I could tell, and 1/3 of the time were completely irrelevant. I have noticed that this ratio appears to have held steady or is in fact getting worse, at least using the major public LLMs. At best, some slopcoders feel that for their applications it may be improving. But in terms of general knowledge even of, say, high school math (and I don't mean asking it to do the equations, just telling me how to do the calculation), they're embarrassingly bad.

I do not see where this "progressing rapidly" stuff is happening. They seem to be churning out the same slop they've always done.

Serious question, but I don't expect you to answer this in a forum like this: do you think it's possible you have a rosy eyed view of LLM tech because it tells you what you want to hear, not just in terms of viewpoints etc, but it keeps reassuring you how right you are, how your ideas are valid, it doesn't question your reason for wanting to do anything, it just supports you, not as a coworker, but as a minion desperate for your approval? And maybe you've never had that from coworkers before, because quite honestly, people in coding especially can be jerks sometimes, and will argue with you about why you're doing whatever you're doing. Perhaps it's possible that after decades of bullying and abuse, because that's what that behavior is even if it's normalized - you're just glad to have someone who does things without being argumentative about it?

There's more to it than just "they're better", there's also massive disparities in labour costs, etc, because of the high cost of living in the US.

If BYD wants to sell cars in the US, why not tell them they can but they need to build them in the US, just as Honda, Toyota, and Nissan do?

The better the tracking data, the more targeted the impression, the higher the clickthru rate. Crafting tracking consent dialogs in a way that favors one app or the other is going to increase the efficiency of campaign placements that target users of that app, rather than campaigns run by app developers who experience a lower percent of users consenting to track.

And they're welcome, as a business to swallow those aquisition expenditures, or fund their own Canadian content, or not do business here.

The list of developed countries that don't have domestic cultural protection policies is low. The US doesn't have one because of its size (or I'll throw you a bone - HOW AWESOME THEY ARE) - it simply doesn't need one.

Kind of hilarious in light of how pro-protectionist things have swung there. If the US audience started consuming predominantly anime TV, movies, books, etc overnight (purely just because of how good they are, right?) you'd bet your ass US would do the exact same thing.

What you don't get is that French is useful for many people - it just wasn't useful for you - and you project that because to you, personal experience is reality. Calling Mandarin, Spanish, Hindi, and Arabic more useful languages for living, identifying with other citizens, and employment in Canada is fucking bananas - and I'm married to a Taiwanese woman and speak some Mandarin!! Nothing you've said has changed my mind. Your points are noted, but still remarkably dumb.

so why aren't they given force adoption

I feel like we need a "take Murdoch5 to a real job" day or something. I'd literally pay to sit you down some time and explain how the real world works.

Supposedly Google's policies are to follow whatever names the locale the user is in uses. In some cases, this is legitimate, as there are several disputed territories around the world and Google isn't a position to pick and choose. On the other, clearly ridiculous situations such as where a President writes an EO, which merely orders his government to do something, and doesn't impact the rest of the country, are an absurd extreme of this.

But, if that's what they want, I'd totally understand if the Canadian government declares all Trump properties in the US having the name "Pedophile Palace", so someone who tries to find Mar-a-lago on Google Maps in Canada is instead presented with that name.

Maybe cheap and petty is the way to deal with cheap and petty.

IIRC it isn't enabled by default, and is considered a dev feature.

The summary and article are, more or less, bullshit.

Basically they're making it sound like these cars are catching fire because ICE cars have some special fault that makes them vulnerable in a way BEs don't. But... turns out all the examples are of the electrical systems of those cars catching fire. Not the engines. The very first example is about the front seat catching fire because the power seat switch on it is poorly designed.

These are all faults that could just as easily be in a BE vehicle. If Kia had a BE version of the Telluride it would have the exact same fault.

This isn't some ICE vs BE thing despite the breathless attempt to suggest it is.

(I say the above reluctantly because I really want BE vehicles to take off, ICE cars just seem more and more ludicrous as time progresses given the massive improvements in battery technologies. But this article isn't the reason to start going "Aha! ICE cars are just as bad!" on that one criticism of BE vehicles that's legitimate.)

Cool story bro.

This article is about Debian allowing the use of genAI for code incorporated into the Debian repos despite the international legal uncertainty, the environmental impacts, and the lack of quality that's largely being ignored solely because software in general has seen a collapse in quality since the early 2010s.

It isn't about your slopware habits.

I'm curious to know whether the people in charge of Debian have a back-up plan if Debian turns out to violate copyright in countries across the world or whether their solution will be to just shrug and say "Well countries that don't allow AI generated slop without compensation to the original copyright holders will just have to do without "open source" - which today is adulterated by shitty unreadable code from slopgobblers."

AI can destroy jobs by destroying the businesses that have inexplicitly decided to rely on it.

Of course, Gates is far too part of the tech establishment in this country to actually say that.

TBH while Trump is pretty much telegraphing that he'll do just that, this story is unlikely to have anything to do with it. We happen to be in an era in which nation states have learned how easy it is to hide secret remote controls in all kinds of apparently innocuous hardware.

It's just difficult with Trump who swings wildly between ludicrous justifications of import bans, and occasionally doing what his administration's smarter advisors are actually telling him needs to be done. One of the reasons this presidency sucks. That and all the fascism.

No, Fox News has nothing to do with the CDA, which was largely about that part of the media regulated by the FCC. Fox News is not regulated by the FCC, it's a cable channel.

I'm not saying the CDA was good - just look at the degree to which radio got enshittified, and local news veered heavily right wing as companies like Sinclair bought up local broadcasters. I'm just saying Rupert Murdoch would have done everything he did anyway as he wasn't affected. The only way I can think of that he benefited from the CDA was 20 years later, when he made a bucket of cash selling Fox Broadcasting to Disney.

To be honest, I don't know why we stopped, because it's not as if modern frameworks take less time to learn or are significantly easier to visualize HTML in.

And many of the sites that migrated to these absurdly heavy web frameworks seem to be just as slow as they ever were. T-Mobile's website comes to mind. Clumsy, falls apart, and makes the fans whir in every laptop I've ever seen - and it's been like this for at least 10 years.

What are devs trying to achieve here when they use them? 'cos it sure as hell isn't reliability, maintainability, or speed. Is it just to narrow the number of people who can replace them?

we're-all-trying-to-find-the-guy-that-did-this.gif

"Just let me live the life I want to live, by fucking up others' lives way more than the other party may have mildly inconvenienced me for sound policy. Wait, why is this affecting me now?"

It's a tale as old as time.