TazeTSchnitzel

We're currently in the middle of an AI goldrush. There are probably hundreds of companies trying to do model pretraining. Anthropic and OpenAI and so on probably care about optics and implementing their crawlers properly. But there will be many firms few of us have heard of. I would guess some of them are Chinese, for example.

This comes off as comically evil. Disabling the AI summaries in search results is trivial, so them refusing to do so and blocking search entirely feels like a blatant attempt at coercion. Isn't it an abuse of monopoly power?

Duplicate of https://news.ycombinator.com/item?id=50314699 (1 day ago, 354 points, 184 comments)

And it is linked to here: https://code.claude.com/docs/en/vs-code (or, well, the URI doesn't tell us if it's using a proxied version or not, so it's complicated)

What software is this? A value-add nobody uses, or a core CUDA component everyone is stuck with?

> CSM here stands for cascaded shadow maps, the shadow technique used by 06.

I was curious about the technique, and there's surprisingly detailed documentation here from MSDN of all places: https://learn.microsoft.com/en-us/windows/win32/dxtecharts/c...

Does anyone understand what's going on here? They expect some kind of request to be signed with a certain key, but what is that request and what does it get you?

The new API confuses me because I can't see what it can really do that you couldn't already achieve by adding some position:absolute elements that float above the canvas. I mean, yes, since it's rendered as a texture, you can apply an arbitrary visual transform… but apparently the input handling is outside your control, so what's even the point? You won't be able to interact with the element properly. I also assume you can't even really occlude things properly? It doesn't make any sense to me.

As a point of comparison: Win32 has APIs that let you draw things like buttons and text input fields without having to instantiate windows (~= DOM elements) for them. Though I'm not quite sure what you can do in the way of input handling without having an actual window. Regardless, browsers manage to contort native controls into funny shapes on behalf of web pages, so the API there definitely offers more possibilities.

Apparently Skatteverket was just about to migrate to Office 365 for email until other administrative agencies complained: https://www.realtid.se/it-tech/skatteverkets-microsoftmejl-s...

I can't fathom why they would willingly try to migrate such critical operations away from a secure state-run system. I'm sure that system sucks in some ways. But those issues can be worked on, surely.

I think it's a hybrid text. The section about how elliptic curve keys work is particularly aggravating. Unfortunately it's also the only section that gave me new information.

> The 46.7 KiB program is the baseline, not the boundary.

what does this even mean

Could HN start feeding submissions through Pangram so us human readers don't have to suffer with reading the nauseating AI-prose?

Oh yeah, there's a handful of words that I learned really early in Swedish that I always get the pronunciation of slightly wrong, even though I don't make that mistake in other words with the same patterns, because it's just so ingrained. And that's in a language with an alphabet!

I think this is a rare “useful” abuse of font ligatures, and in that respect I think this is a cool project, but it's also a terrible idea that nobody should actually use, because Japanese kanji readings are famously ambiguous and context-dependent. The modal character has two different readings without counting those used in personal names, nor contextual slight changes to pronunciation. Some characters have six different readings, and that's again without counting those used in personal names. There's no way a font can do a good job at it unless you can somehow stuff an entire LLM in there.

For context, it's a hard enough problem that Google Translate's own romanised pronunciation guides are frequently wrong, or disagree with their text-to-speech.

The original Half-Life also supports software rendering, and it has a unique water texture effect that the OpenGL renderer doesn't have.

Do the kids not know about about:blank?

Blocking the payment doesn't solve the problem, you may avoid your bank account being instantly emptied but you're still on the hook for the money.

The details of this are so absurd that it took me quite a while to realise this was talking about real events and not some kind of satire.

Is it maybe providing a bogus NTP server or something? Maybe the automatic reboot feature can be moved to the Secure Enclave or something, and made to only rely on the hardware RTC in a way that can't be tampered with.

This sounds a bit like Smalltalk! But with partial application…

It also reminds me of the Emily programming language: https://github.com/mcclure/emily/blob/stable/doc/tutorial.md

I guess that demonstrates the concept of “a drop in the ocean”, but turning a massacre of civilians into a unit of energy feels crass at best.

> The fix is to set bit 13 of MCSR24 to 1. MCSR24 is an internal CSR whose function is not described in the manual. After setting this bit, the lost update no longer occurs. Testing showed the performance loss is very small: single-core performance is unaffected, and multi-core performance drops only slightly.

Presumably this is a so-called “chicken bit”.

“Clan voting” and “mosque voting” are terms that try to exoticise the phenomenon with a racial overtone. I don't think they are conducive to a healthy discussion. There's an in-built assumption in those terms that these voters are only voting a certain way due to some kind of special “foreign” type of social relation that Swedes without immigrant backgrounds would not have. This plays into prejudice and is not necessarily supported by the evidence. It also presupposes a particular conclusion.

I think you can be concerned about voters with low levels of education and poor socio-economic circumstances being pressured to vote one way or another without immediately framing it in this racialised way. Why must it be a “clan”? Does such a concept even have a clear definition? Wouldn't Swedes with a Somali background perhaps be inclined to support someone else who is also Somali if they happened to be a candidate for political office, as they would feel represented by them? How do we know they are from the same “clan” and not just the same country? Are Sweden Finns who vote for Sweden Finnish candidates “clan voters”?

Similarly I can point out that mosques are not special. If someone is influenced in how they choose to vote by what they hear at church, are they a “church voter”? Remember that religious institutions are often at the centre of social life in certain communities.

It looks like the OP is determined to push a certain narrative about election fraud being pivotal in the 2026 Swedish election.

I think in that context it should be pointed out that the 2026 election was not particularly remarkable. It went from a 2-seat majority for the liberal-conservative bloc to a 2-seat majority for the liberal-left bloc (there's two different liberal centrist parties, one in each bloc… Sweden's a funny place). There was a mere 0.67 percentage point increase in turnout from last election. Some of that may be from recently naturalised citizens, who according to polling are somewhat more likely to vote for the left-leaning parties (not strange given those parties are less hostile to immigrants). But the biggest change in this election is the 3.05 percentage point drop in support for the most right-wing anti-immigrant party. I don't think you need to resort to accusations of fraud to explain that one: the anti-establishment populists have been in power for the first time ever, and been forced to compromise. It's not surprising that voters become a bit less enthusiastic. But only marginally: this was an election won on a single-percentage-point swing. If there's been large-scale fraud here, it's very very subtle.

Whoa there, hold your horses:

- There's an ongoing investigation into alleged election fraud, which has not yet proven anything

- The candidate himself is not even a suspect in that investigation

- The candidate is known to have had some contacts with Somali government officials in the past, but no relation to this election is alleged by said investigation

- So far, it doesn't sound like they even have actual evidence any fraud took place (https://www.svt.se/nyheter/inrikes/senaste-nytt-om-val-2026?...)

I believe in innocent until proven guilty. Remember that political opponents might have an incentive to fabricate allegations.

Sweden uses a party list proportional representation system. Seats are allocated based on how many votes a party gets, and they are allocated to parties first. The selection of who then sits in those seats then comes down to ”personröster” (literally “person-votes”, but I think they're conventionally translated as “preferential votes” or something like that), which are optional. Most voters do not put a cross next to their preferred candidate, they just put the party ballot paper in the envelope and the envelope goes in the box. And because they're optional, there's therefore a threshold for how many personröster a candidate has to receive to be elected. That threshold is 5% of voters in the particular electoral district. Only candidates that get more than 5% of votes are even considered in this process; if there's none (or not enough), then the candidates from the party list are taken.

The seat would have gone to the Left Party either way. But only this guy specifically got enough personal support to get elected. There's nothing that crazy about it. Participative democracy rewards those who choose to participate. In this case, it's a two-tier participation system. People who vote for a party, but not for a specific candidate, are letting the people who do vote for a specific candidate choose for them. And that's fine.

This article from back in March correctly predicted that the problem was, essentially, Palantir: https://www.theguardian.com/news/2026/mar/26/ai-got-the-blam...

> Although not perfect, [Sound Canvas VA] is the closest recreation of the original SC-55 in software form as of 2023.

Ah, this submission should have a “(2023)” added to the title. In 2024, Nuked-SC55 (https://github.com/nukeykt/Nuked-SC55) was released, which is a fully accurate software emulation of the SC-55/SC-55mkII (it relies on having the original ROM files).

FrontPage didn't generate your content for you! So there was no reason to assume a FrontPage page might be meaningless just from how it looked. For AI-generated sites, though, if the page looks like slop, is it worth bothering to figure out if it actually is slop? It's not like they're uncorrelated variables. Sure, occasionally it won't be. But I'm drowning here. I must simply assume that if someone is serious about conveying a message, they won't package it in this form.

I don't think it's even gatekeeping. You could copy-paste two tables and write two sentences into a Word document and export to HTML, one of the laziest possible ways to produce a website, and it'd still be less off-putting because it doesn't have the association with slop content.

You could be presenting the most carefully-researched information in the world, but with such an obviously vibecoded site I'm not going to attempt to read it.