paimapi

From your own link, of all the EU nation states Spain is pretty much middle-of-the-pack for per-worker hours.

Another counterpoint: higher work hour countries have higher rates of death per kilometer of active track [0]

Greece (highest in the EU for work hours) = 22.3 rail deaths per 1k kilometer of track followed by Portugal (another high work hour ranking) with 6.2, Portugal (about 200 more hours than Spain annual) at 6.2, and Hungary (at about 100 more hours) at 5.8.

The only two countries with less deaths per 1K KM of track and more work hours are Ireland and Estonia. All the rest (Denmark, Sweden, Austria, Finland, Slovenia, Luxembourg) have less annual work hours than Spain.

[0] https://ec.europa.eu/eurostat/web/products-eurostat-news/w/d...

Help me understand a historical period when this was ever the case.

Social historians have been debunking the thought-terminating cliche that we ever had a trend of merit-based hiring since time immemorial:

https://thereader.mitpress.mit.edu/the-myth-of-meritocracy-r...

https://archive.org/details/meritocracymyth0000mcna_n6g9

https://historicalstudiesineducation.ca/index.php/edu_hse-rh...

Some questions from a netsec novice:

1) Why wouldn't you have a SSOT for certs and instead rely on patching different apps individually?

2) Doesn't this violate defense in depth? Exploits for a widely disclosed issue now have turnaround activation within hours, not weeks. Not patching all surface areas and actively disclosing as such seems very much like drawing up a map for a jailbroken LLM to gin up a rudimentary attack.

3) Also doesn't this violate the principle of secure by default especially for power users who are more likely to be the ones with higher privileges across distributed systems?

How do you even speak to the ineffable idea of 'incompetence'? How do you remediate it? More draconian, individualized worker surveillance?

This feeds into the idea that the problem with safety these days are lazy workers, not the fact that companies are increasingly extending work hours beyond what's reasonable, firing keystone employees and shoveling more work onto the remaining individuals, and operating with a general disregard of regulatory safety guidelines around training and rest.

The reality is very different: https://www.railway.supply/renfe-rejects-drivers-union-claim...

> On rolling stock, the union wants lasting technical fixes for reliability. It further calls for correct application of breaks in driving time under article 36 of Royal Decree 929/2020 and an end to the block on early-retirement reduction coefficients under Royal Decree 2621/1986. Two further demands concern a May 7 agreement on hiring new entrants as trainee drivers and a 1990 agreement on support for drivers involved in incidents where a person is hit by a train.

>On new drivers, Renfe called the claim that recruits are being kept on training scholarships false. It said none of the 402 drivers who have joined are on a scholarship.

And they have: https://www.wsj.com/business/telecom/rocket-lab-buys-satelli...

To be fair, this was obvious to many people who study domestic civil rights as early as the bill that originally created the DHS and ICE. For eg, this review from 2002:

https://digitalcommons.wcl.american.edu/cgi/viewcontent.cgi?...

>Americans’ liberties have been trammeled in a variety of different ways. Under the guise of stopping terrorism, law enforcement officials and government leaders have now been given the right to conduct searches of homes and offices without prior notice, use roving wiretaps to listen in on telephone conversations, and monitor computers and e-mail messages, even to the degree of eavesdropping on attorney/client conversations.

>Conceivably, these extensions of the definition of “terrorist” could bring within their sweep diverse domestic political groups, which have been accused of acts of intimidation or property damage such as Act Up, People for the Ethical Treatment of Animals (PETA), Operation Rescue, and the Vieques demonstrators.

It should have been stopped before it grew into the leviathan it is now but the general consensus by the politicos of the time was that this was just histrionic left/libertarian worry, and that something like that could never happen in the USA, we were too just, too democracy-loving.

What about "it's not Fly.io and it's more about how the Fourth Amendment is effectively neutered in our dystopian, socially regressive era."

You think a company can exist with just executive leadership?

Would love to see just one example of a billion+ valuation company with nothing but C-suite and that against the ratio of companies with 1000+ employees who are worth a billion+. The only two sub-dozen workers that come to mind for me are Theranos and Bernie Madoff and I'm not sure they are examples of success.

2018-2022 was also when interest rates were at historical lows so companies were placing 'big bets' with less-but-not-negligible risks. Same was true for hiring too - remote work + lots of liquidity meant a massive hiring spree that we're now, with our 5x higher interest rates, very much in the comedown for (and which many leaders pin on 'AI-assisted work' and not on their really rash and stupid bets not panning out).

That seems really great, though. If you showed up with an ancient retro PC, you're asking for other nerds to chat you up and ask you what you're tinkering with.

Nobody's going to want to talk to you if you show up with a full iMac setup.

To a certain extent, yes. But you could apply that same logic to people who do indecent exposure. At some point, if you're setting up for 4+ hours in a tiny coffee shop with a laptop replacement and you're not buying more than just a cup of joe, you're doing a net-bad for your community by making that coffee shop and thus cheap community spaces less viable.

If you want to do it in a Panera or SBux go ahead but not in a local shop. Being conscientious is actually a good thing, believe it or not.

Do... people drag a full iMac to a public workspace like that? I feel like this would be ridiculed since even most co-working spaces provide monitor hookups. But if I saw you order a single coffee at a tiny shop and are set-up with an all-in-one desktop then me and everyone else will be giving you a side-eye

Slightly less worse would be one of those fold-out multi-monitor attachments that I've never seen in the wild either

Why hire crew or actors when you could have them be fully AI? Why film a commercial with fluffed, oil dusted, well-lit hamburgers when you could fully render something? Why actually have a vehicle capable of off-roading when you could just AI-render what it looks like off-roading (or vertically taking off, for eg)?

It's a dystopian business use case that I hope will get obliterated in the near future as the generations that believe in this stuff and are fascinated by it age out of the population.

Thanks, Obama! (literally - https://www.techdirt.com/2013/03/07/holder-doj-used-discreti...)

I prefer Graeber's definition (RIP) -

https://theanarchistlibrary.org/library/david-graeber-neolib...

So what is your preferred noun phrase?

This is pedantic but that's not proper usage of 'shibboleth.' The basic conceit of an 'AI booster' is easily defined by the word 'booster' itself - it's someone who 'boosts' AI, likely without much actual subject matter expertise on the matter and the word 'booster' has a long etymology applied to various ideas: https://www.etymonline.com/word/booster

When most theorists talk about shibboleths and impenetrable academic writing, it's in reference to terms like 'deterritorialization' or 'intertextuality' which are not immediately evident to most lay people who aren't armed with an etymological background or who once stumbled into an advanced course on literary theory in pursuit of a creative writing degree

You could probably say that there's cultural capital associated with it but that's true of nearly any term describing an ideology or opinion. The very same AI boosters are guilty of coining things like 'orthogonality' or 'instrumentally convergent subgoals' which lean more towards being unnecessarily bombastic than descriptive (I guess 'intelligent but amoral' and 'self-preservation' aren't fancy enough for adoption) which loses the strongest argument for academic jargon like the one advanced by Butler: https://web.archive.org/web/20240630040227/https://www.nytim...

Arguably, I would say this has been a 50+ year-long project starting in the Reagan/Thatcher era that Clinton/Blair fully realized by ushering in de-regulatory, labor-disempowering neoliberal policies as part of the left political flank. There's been a lot written about this:

https://jacobin.com/2026/09/labour-callaghan-neoliberalism-m...

https://dissentmagazine.org/article/the-legacy-of-the-clinto...

https://sk.sagepub.com/hnbk/edvol/the-sage-handbook-of-neoli...

Did we read different articles? None of the issues they brought up (architecture splitting, memory transport, Cloudflare's misconfigured context window provisioning, the SGLang state leak, lack of client error provisioning) are model related. These all exist at the layer of the harness (ie Cloudflare's configuration).

A better title would be "Cross-request KV cache exposure in Cloudflare provisioned GLM-5.3 Workers AI model" because it narrows culpability. Just posting it on the Cloudflare blog isn't enough for that in the age of AI and SEO data-scrapes and it'd be surprising to me if the authors didn't know that.

It's fairly widespread and has been normalized for years. A very quick internet search turned up this:

https://www.nytimes.com/2024/07/13/business/dealbook/ai-prod...

https://onlinelibrary.wiley.com/doi/pdf/10.1111/newe.12400

https://www.technologyreview.com/2025/12/23/1130393/how-soci...

Right. People don't understand the difference between labor and value. The job of a corporation is to extract as much surplus value from its workers as possible while making it as opaque as possible to individual workers the actual revenue they bring in.

It's very normal to think that the extra revenue growth you bring in should mean you have either 1) a larger share of the pie for every efficiency increase that grows revenue or 2) less expectation from you, as a worker, to labor as intensively since you are continuing to bring in surplus value that is being actively skimmed from.

This is why I hope more people contribute to the AdNauseum project. Trying to limit cookies and trackers is a losing game. We need to proactively add buckets of junk data for advertisers - that's the only way to get them to stop collecting so much.

https://adnauseam.io/

Not to mention the long history of whistleblowers.

Lots of lists of people who, in a just society, would be valorized and treated as national heroes:

https://www.whistleblowers.org/how-whistleblowers-changed-th...

https://stfrancislaw.com/blog/10-most-famous-whistleblowers/

Years of corporation-favoring jurisprudence. Our laws protect the interests of capital, not human rights.

A wise man once said, "Laws are threats made by the dominant socioeconomic-ethnic group in a given nation. It's just the promise of violence that's enacted, and the police are basically an occupying army."

See:

https://www.eff.org/deeplinks/2021/07/supreme-court-narrows-...

https://news.bloomberglaw.com/us-law-week/u-s-supreme-court-...

Not in America, no - https://eji.org/issues/qualified-immunity/

We need our slave catchers and Pinkertons as immune as possible from accountability. How else will they protect the interests of capital?

Pangram v4 shows 100% human written: https://www.pangram.com/history/aa7e5cb1-0070-4bcc-8694-6180...

Don't let AI tells ruin professional writing for you.

I would imagine that using an LLM model for secondhand thinking would be seen as a bit decadent and reflects a weak will, easily influenced by the illusion of authority.

The Will to Power, as far as I understand it, is about how resistance (or friction) produces the strength of will to overcome obstacles. AI does the virtual opposite of that while also commoditizing cognition.

I'd agree with the other commenter that Nietzsche would probably see this as just another reversion to Plato's Forms - LLMs, by nature, are median outputs of cognition. There is no ubermensch to be found amongst the lowest-common-denominator, only the narcotic allure of becoming increasingly, as the kids say, smooth-brained.

I would maybe argue that the vast majority of science reporting especially in large news outlets is bad science reporting since the modern media ecosystem has strong preference for sensationalism.

It's why there's so much distrust of science globally. We've cured cancer! Except we haven't, it was mouse studies. Except we have! Except not really, that drug failed it's first clinical trial. Except we have! Except that's just mechanistic reasoning not proven by any observational studies.

So then people think, wow, these scientists have no idea what they're doing and just make stuff up all the time to get attention. Which they aren't - most peer-reviewed studies include a mountain of caveats and I've rarely seen, even in large meta-analyses and Cochrane Reviews, papers that make half as strong a claim as ad-revenue obsessed news editors do with their hyperbolic headlines.

They are effectively offering a user's data for sale, just obfuscated with their internal translation layers with a modicum of PII stripped. I'm calling you pedantic because you're conflating intent with impact. The impact of Google's advertising platform is hyper-narrow targeting of individuals based on their PII for a price.

Plus, this service requires that you trust Google to store that data which, for the third time, is not exactly great given their collusion with agencies like ICE (something you've ignored repeatedly now). In the age of NSPM-7, this is absolutely a concern worth agitating about: https://www.aclu.org/news/national-security/how-nspm-7-seeks...

Also, this service requires you to trust that Google spends enough money preventing intrusions which, if you've been following infosec news, is no longer even a modicum of a priority for them: https://x.com/loobeny/status/2049942398416122250

You had a strong reaction. Now you're digging a hole, ignoring points that detract from your reactionary thesis, and ignoring the wider context in order to focus on an accusation fully based on your specific interpretation of my writing, not the substance of the content. If you want rational discourse then you should do better than simple sophistry - just because you have some pet peeve that you globally accuse all HN commenters of having doesn't make it more true. Those who claim they have no confirmation bias are the ones who are the most likely to have it.

I'm not sure why you're going so hard in the paint to be pedantic but sure, Google doesn't literally sell their exact byte-for-byte data, what they sell is a service that is functionally a few translation layers removed from their actual database. This is pretty much how all advertising SaaS works. When I wrote my comment, I was writing to an audience that I assumed was well aware of this given that most of us work in tech.

You can use that service to get enough context clues about certain targets to identify someone [0] [1]. Will most advertisers do this? Not five years ago but it's vanishingly easy now to get an LLM to gin up bespoke algorithms that will do it for you. And the door between Google's servers and federal LEOs has no latch and they share plenty of byte-for-byte data with agencies like ICE (which is literally one of my links).

I don't understand the point of being so hyperfixated on word choice since this is just Hacker News and not a court of law. But I suppose everyone makes their choices for a reason even if that reason has more to do with some reactionary and emotional outburst rather than any actual substantive additions to the discussion.

[0] https://arxiv.org/abs/1510.04031

[1] https://www.wired.com/story/google-dv360-banned-audience-seg...