Lvxferre

What about the bean bags and snacks? Lust isn’t the only deadly sin dammit, you got to think about sloth and gluttony!

I’ve eaten pigeons plenty times in my childhood. That my grandma “recovered”. But to be fair back then my neighbourhood was suburban-ish.

You’re right. And IMO they should be legally banned from doing so - because the people who signed up for this crap agreed with 23 and Me’s ToS, not with someone else’s.

But, well… as you said, capitalism going to capitalism. The “right thing to do” is often out of the table of options.

I have a relative who considered doing this test. I’m glad that the family talked him out of it. (Surprisingly enough, not just me.)

Anyway, my [hopefully not “hot”] take: for most part the data should be destroyed, as it involves private matters. If there’s data that cannot be reasonably associated with an individual or well-defined group of individuals, perhaps it could be released into the public domain, but I’m not sure on that.

That’s a better way, I agree.

The fun part isn’t even what Apple said - that the emperor is naked - but why it’s doing it. It’s nice bullet against all four of its GAFAM competitors.

Your donuts are more like don’ts. I invented better donuts. Here they are:

I shall call them Poppy and Betty. [WARNING: THOSE ARE OC, DONUT STEEL.]

It’ll likely turn out that the more dispassionate people in the middle, who are neither strongly for nor against it, will be the ones who had the most accurate view on it.

I believe that some of the people in the middle will have more accurate views on the subject, indeed. However, note that there are multiple ways to be in the “middle ground”, and some are sillier than the extremes.

For example, consider the following views:

1. That LLMs are genuinely intelligent, but useless.
2. That LLMs are dumb, but useful.

Both positions are middle grounds - and yet they can’t be accurate at the same time.

Here’s a simple test showing lack of logic skills of LLM-based chatbots.

1. Pick some public figure (politician, celebrity, etc.), whose parents are known by name, but not themselves public figures.
2. Ask the bot of your choice “who is the [father|mother] of [public person]?”, to check if the bot contains such piece of info.
3. If the bot contains such piece of info, start a new chat.
4. In the new chat, ask the opposite question - “who is the [son|daughter] of [parent mentioned in the previous answer]?”. And watch the bot losing its shit.

I’ll exemplify it with ChatGPT-4o (as provided by DDG) and Katy Perry (parents: Mary Christine and Maurice Hudson).

Note that step #3 is not optional. You must start a new chat; plenty bots are able to retrieve tokens from their previous output within the same chat, and that would stain the test.

Failure to consistently output correct information shows that those bots are unable to perform simple logic operations like “if A is the parent of B, then B is the child of A”.

I’ll also pre-emptively address some ad hoc idiocy that I’ve seen sealions lacking basic reading comprehension (i.e. the sort of people who claims that those systems are able to reason) using against this test:

• “Ackshyually the bot is forgerring it and then reminring it. Just like hoominz” - cut off the crap.
• “Ackshyually you wouldn’t remember things from different conversations.” - cut off the crap.
• [Repeats the test while disingenuously = idiotically omitting step 3] - congrats for proving that there’s a context window and nothing else, you muppet.
• “You can’t prove that it is not smart” - inversion of the burden of the proof. You can’t prove that your mum didn’t get syphilis by sharing a cactus-shaped dildo with Hitler.

I still fail to see how people expect LLMs to reason. It’s like expecting a slice of pizza to reason. That’s just not what it does.

This text provides a rather good analogy between people who think that LLMs reason and people who believe in mentalists.

Then the kid installs Proton and your keikaku* fails miserably.

*TN: keikaku means plan.

As I mentioned in another thread, about the same topic:

First Zendesk dismissed the report. Then as hackermondev (the hunter) contacted Zendesk’s customers, the issue “magically” becomes relevant again, so they reopen the report and boss the hunter around to not disclose it with the affected parties.

Hackermondev did the morally right thing - from his PoV it was clear that Zendesk wasn’t giving a flying fuck, so he contacted the affected parties.

All this “ackshyually it falls outside the scope of the hunt” boils down to a “not our problem lol”. When you know that your services/goods have a flaw caused by a third party not doing the right thing (mail servers not dropping spoofed mails), and you can reasonably solve the flaw through your craft, not doing so is irresponsible. Doubly true if it the flaw is related to security, as in this case.

I’m glad that Zendesk likely lost way more than the 2k that they would’ve paid hackermondev for the hunt. And also that hackermondev got many times over that value from the affected companies.

It needs to work and be reliable

True that. And it needs to offer a better service than Google; “our services are federated so please use them!” is not enough to encourage the shift, I think.

The video can be summarised into three main points:

1. Advertisement offering Google a perverse incentive to make its search results worse, so the search ad results look comparatively better.
2. Search engine optimisation.
3. Generative AI integration with Google enshittifying the platform.

I’ll focus on #2. Federated search might alleviate the problem.

It’s counter-productive to optimise a page for multiple search engines, running different algorithms; it might perform better on [let’s say] Google, but worse on [let’s say] Bing, or vice versa, since they run different algos that prioritise different things. As such, almost all SEO is made for Google results.

And, in an environment where no search engine dominates the market, and the search engines use different algos, SEO goes away.

The problem with that is people don’t want to use multiple search engines - they want to use one, that they believe to bring the best results on. (That’s why we have a problem called Google on first place.) If only there was some way for those search engines to coexist, and to benefit from each other… well, that’s basically federation, right?

How I see it working:

• each instance crawls the web separately, focusing on the pages that it wants to
• each instance has its own ranking algorithm
• each pair of instances may opt to federate with each other or not
• each instance can relay search queries to each other, if they’re federated
• as a user inputs a search query, based on keywords and/or user preferences, the instance might decide if it should service the user with local results (from that instance), with results from a federated instance, or a mix of both.

I believe that this system would make SEO really hard to do; in practice you’d be better focusing on good content. It would also lead to a situation where different search engines want to specialise, but still keep each other alive - as they benefit from their peers.

Source.

It’s in Italian. I should’ve changed it before the screenshot, but it takes effort to do so, and I’m lazy.

My computer settings are all in Italian, to avoid getting rusty with the language. Sabe como é - não usa, enferruja.

He’s a jumento / donkey and a piece of shit on moral and ethical grounds; he doesn’t give a flying fuck about the population, and seems to have a burning hate against marginalised groups. However, he isn’t too prone to shoot his own feet.

This is relevant in this case because social media is essential for what he’s doing: gathering support for his meat puppet’s potential election, while playing the victim of an anti-democratic government. “Poor me, I was unjustly prevented from being a candidate! The powers to be don’t want me to! Vote on $person by the way.” As such his interests align really well with Musk.

My hypothesis:

Meta is one of those companies wallowing in the idiotic belief that generative AI will “soon” reach intelligence and sentience and the ability to walk your dog, so odds are that it’s deploying them heavily for moderation duties. Except that the crap does not understand a single iot of the pictures and text that it analyses, so it’s bound to get huge amounts of false positives and false negatives.

Well, here’s an example of false positive. i.e. machine mod assuming that the poster is underageb&.

Protip: if you use “assumer machine” to handle people, you’re trash, your service is trash, and you both deserve to be treated as trash. Not this conclusion is surprising regarding Meta.

Would

Fight it? Bad idea. That’s the wandering spider’s way to say “back off! I’m going to jump on your face, bite you, and you’ll die an agonising death! I’m serious!!”

[Still cute though. At least from a distance.]

Can confirm that it’s on again:

If the Brazilian government had actual laws against the issue with Xitter /'ʃɪtə/ - i.e. the fact that it’s a Nazi nest - it wouldn’t be on again. However Moraes did it through some bloody convoluted way, that stinks “ackshyually” from a distance; first obsessing over that Monark clown, then the representative, then a fine that amounts to pocket money for both sides. As a result, it’s on again, “yay” /s

The key here is that legislations (not just the Brazilian one; I’m talking on general grounds) need to distinguish between

• a platform that transparently conveys the discourse of its users, thus not being responsible for what they say; and
• a platform that opaquely “suggests” you what you “should” be seeing, through an algorithm that you cannot reasonably modify, thus being liable for what it shows you

Faecesbook and YouTube fall into the later, but Xitter specially so, as Elon Muppet’s personal soapbox.

So, let’s give a round of claps to Dickhead for the Effort and Good Intentions® /s.

“By ‘free speech,’ I simply mean that which matches the law. I am against censorship that goes far beyond the law,”

Cut off the crap. You were probably babbling this on the grounds of that “first amendment” crap that holds no grounds outside USA.

Bolsonaro and his supporters praised Musk for refusing to block accounts

Of course he did. He might not be himself eligible any more, but odds are that whoever will compete for the Republic’s Party Liberal Party in 2026 will be basically his meatpuppet. People often compare this shite with Trump, but they forget to mention that, unlike Trump, Bolsonaro is not braindead.

President Luiz Inacio Lula da Silva previously said that businesses in Brazil must follow local laws and that the country is “not obliged to put up with Musk’s far-right ideology just because he is rich.”

Lula, you’re only there because you’re less worse than Bostonaro. Now go drink some booze and shut up.