miki123211

The case centered on the copyright battle involving Anne Frank's diary

If there's a copyright dispute somewhere, you can bet it's either Anne Frank's estate, Tolkien's estate, or the Beatles's old record label.

It's never the authors who actually made the work. It's always their heirs.

CSA doesn't neatly fit into a narrative.

If there's a person committing abuse against a child, usually a family member in their own household, there's no story there. They (hopefully) get arrested, get a minor mention in the media if that, and life moves on.

If, on the other hand, they send CSAM using a platform, device or protocol, then clearly the makers of that technology failed to uphold their sacred duty to protect the child. If most CSAM cases happen using a few major technologies (and they will, because societies tend to standardize on what technologies they use for communication), the technologies get blamed for the problem they supposedly enable.

If you dislike the technology for other reasons (the distrust of corporations from left-leaning politicians, censorship allegations on the right, a moral panic about the impact of smartphones on children's mental health), you can use people's misunderstanding of statistics to exaggerate the CSAM problem, blame the tech, and gain some notion of control over it in a way that is politically palatable to citizens.

It's worth saying explicitly that centralization doesn't have much to do with this. We've seen similar stories play out with bicycles, Walkmans, pagers, AI, heavy metal and Uber rapes. They were different moral panics, but the mechanic was roughly the same. I think the situation wouldn't change much if we all used PGP-encrypted email over personal mail servers to communicate.

An exception that proves the rule is the moral panic over CSA in primarily-catholic countries, notably relating to abuse committed by catholic priests, which are statistically no more likely to commit it than anybody else. As the influence of the Church on government policy is a hotly-debated topic in those countries, CSA is suddenly an issue that people can use to further their political causes.

researchers found that more than 42 percent of games and other apps available for download on LG’s webOS store allow unknown third-parties to route their Internet traffic through a user’s TV.

Eh, so even sources as reliable as Krebs on Security sometimes engage in bad statistics.

This is a statistic that feels significant, but actually tells you nothing. For all we know, "42% of all apps" can mean a single spray-and-pray campaign with a few hundred downloads in total, with the vast majority of users only downloading apps from the top10, which are all unaffected. Or maybe the problem is widespread across the entire distribution of app downloads, with the majority of users having at least one infected app. We don't know either way, because "42% of apps" is a completely meaningless statistic.

From what I've heard, Canada is actually a lot worse than the US.

I was traveling as part of a party of five (different travel dates, separate ETA filings, no apparent connection). One of us didn't get an ETA. Another got one just before the trip (and it was a 13-year-old child). The rest of us got ours immediately, but I was questioned at the border for 10+ minutes with no apparent reason, and then allowed in without a comment.

Yes.

In fact, you have to do it twice; once to apply and once to receive it.

In Poland at least, the queues are often a few weeks long, and you have to book online in advance. If it's summer season and you need to travel soon, you usually have to go to some office in Buttfuckville at the other end of the country to get a reasonable date. Passport office hunting is an established practice for people who need one ASAP.

If you're a kid, both of your parents need to be present with you at the appointment at the same time, so they have to coordinate time off work. The fun begins if they're divorced and one lives in a different country.

At least, unlike in the UK/US, the process itself is pretty painless. I got mine just after I turned 18 (see above), and all I really needed to do was to go there, bring my ID, fill a single form and give them the fingerprints.

Which you don't need to do, because the fingerprints are in the passport (cryptographically secure so only an authorized reader can access them, and only when able to OCR the inside of the passport). They're signed by your issuing government, so they're much harder to forge than the physical document.

With port forwarding, somebody else still owns the identifier that points to your website, whether that's a domain name or an IP address.

A huge advantage of GH / CF is that your site has a strong chance of outliving you, if that's something you care about.

If you're self hosting, either at home or on a VPS, you have to worry about domains expiring and credit cards being cancelled, which is inevitably going to happen if you don't have a succession plan in place. You could probably solve it with a bunch of lawyering, but that's expensive.

Free services don't have this problem. They can still go away, it's not a 100% guarantee, but it's a "might" versus a "definitely will"

Accessibility.

All accessibility stacks sucks in some respect, but Linux's sucks most of all, and Wayland people in particular don't seem to be willing to compromise on security (which is required for accessibility to work).

For audio interfaces, HDAudio lets you do most of this. What the extra software gives you instead is often some crappy DSP that increases latency and messes with sound quality.

For years, Dell's / Realtek's software had an unpaged memory leak somewhere. If you were using a screen reader (I guess they must interact with audio devices in some very specific way that Realtek hasn't accounted for), your system would eventually run out of RAM and BSOD. They didn't fix this until Microsoft and a few screen reader vendors intervened. "Don't buy Dell" was a standard recommendation in the blind community for years, which didn't help if you had a work PC with no local admin.

8. ANd this isn't specific to LG. If LG can do it, anyone can, even if they aren't right now.

Buying from companies you trust isn't a solution either. Founders sometimes get into fatal car accidents or lose some of their assets in messy divorces. THe new owners may not care about "brand reputation" and sell the company to the highest bidder.

That's called a screenshot.

Google search or a traditional paginated forum doesn't provide that either. No guarantee that page 15 of thread X will contain the same posts as it does now when the moderators wake up and delete the flame war.

There's a spectrum between a restrictive sandbox and full YOLO mode, may 2025 style.

You can have something like Claude's auto approvals. You can poke specific holes in the sandbox (E.G. my Codex one can write to ~/go, ~/.cache and ~/.cargo). You can have explicit deny rules that hard-deny and bypass the sandbox and auto-approval. You can allow certain commands to bypass sandbox execution entirely.

While those aren't secure in theory (if you let a third party type prompts into the agent, they will likely be able to exfiltrate your data), they're secure against agent carelessness and AI stupidity in practice.

Systems that are based on one-time codes instead of permanent credentials are really under-appreciated in the anglosphere.

Giving your credit card to your child / friend / spouse / subordinate, so that they can buy something with your money, is a lot more risky than giving them an OTP, valid for a single transaction, which you have to approve in your app anyway.

Poland's Blik is based on 6-digit codes, which makes this feature even more useful, as you can E.G. dictate that code over the phone to a trusted acquaintance at the ATM.

I see Rust macros as the closest thing that demonstrates homoiconicity in mainstream Algol-based languages

I don't know if you consider Elixir mainstream, but IMO their macro system is much closer to lisp's ideal.

Elixir is basically Lisp, but with better syntax, a modern ecosystem, and running on the Beam. Unlike languages like Rust, Elixir's conditionals and function definitions are just calls in the AST, even though the syntax looks mainstream and not like paren soup.

To be fair, I don't think Discord supports HA. As a pretty active user of voice calls on Discord, I've had plenty of cases where the entire channel drops for a second or two and reconnects, presumably to a different server.

If you do some kind of rolling upgrade, where each server has E.G. a lifetime of two weeks, with a 5 hour window where it doesn't accept new connections, I think you'll be fine.

Oh, but you can PoW every page.

Your typical end user doesn't switch IPs that often, so it's fine to Anubis them again when they do. A scraper, on the other hand, has a tradeoff to make between rotating ips often (requiring a challenge on every request) or keeping only a few IPs (making cross-request identification much more valuable and reliable).

Most CF / Recaptcha problems are users going "off the golden path", and not realizing that their config changes are at fault.

If you're on a consumer router, using a mainstream stock browser with stock settings (maybe plus uBlock Origin), with your Google account logged in, it's very, very likely to just work. If you're part of the .01% of users with opinions about that sort of thing... you're not worth optimizing for.

The reason is UI.

There's only a limited number of features that you can pack into a few buttons and a 7-segment display. If you want to sell outside the US and need to support the long-tail of non-English languages, preferably without per-country product variants, you can't even label the buttons any more, you have to rely on simple pictograms and icons.

If there's a $1 microcontroller in your device (and there often is), you're very tempted to implement lots of features which cost you almost nothing, but that kind of UI just doesn't really let you do so. Sure, you could add a proper touch screen with a localizable UI stack, with reflowable text and support for displaying Kanji and RTL languages, but that's often more expensive (and less practical) than slapping on a BLE or WiFi chip.

The hard part isn't controlling the fan, it's controlling the fan from anywhere, without a central server in the way.

Sure, you could probably make this much smaller if you invented a specialized p2p fan control protocol, but that's a lot of work.

Once this gets video capabilities and is ported to glasses, it'll be a major revolution for blind people (and I say this as a blind person).

People have tried "smart <thing> that helps blind people navigate" since the 80s, many, many, many times, and all such projects failed. The cycle of "wow, blind people could benefit from a navigation aid, why don't I make one, if there's none around, I must surely have been the first bright university student to think of this idea" is pretty well known in the community, and I'm personally quite tired of it. Nevertheless, I think this may be the one.

Circa 2020, I have said that people who are getting a guide dog now are probably getting their last one. I think we aren't far off from that prediction coming true.

I wonder how they'll handle APNS, which will have to happen eventually.

Doing it costs money, and it's not a cost you can simply shift to the end user (because only the original app developer has the required Apple certificates, and you don't want every server owner to sign up for the Apple Developer program)

Your getting started docs are extremely confusing.

Sure, the docs tell you exactly what to do to start a server, but not how to sign in to it. Or how to sign up (email is disabled).

There's an `operator` command which is supposed to let you create users, but it can't find the operator API.

You were that close to a perfect onboarding experience...

In my experience, all Anthropic models tend to produce hard-to-understand word soup. They're like the kind of person who likes to use fancy words just to show you how much smarter they are.

That doesn't mean those models are stupid or just pretending. I do think that Fable is the best we can get right now. Nevertheless, I prefer OpenAI models for easier tasks where the primary output is text and explanations, not code.

Anthropic and OpenAI are really two different businesses in one company.

One business is serving great models via the API. I think that one is indeed going to become commoditized. The other is making consumer / developer focused products like Chat GPT, Claude Code or Codex.

Anthropic is leaning hard into making Claude Code work with things like enterprise compliance policies. I wouldn't be surprised if at least one of these companies basically ends up as a SaaS fronting many different models for different capability points, some of which are open.

How do you distribute credentials to residents?

My Macbook is permanently locked out of Cox's hotspot system (used in some U.S. hotels) because the password was given to me on a tiny label which I couldn't read as a blind person except through OCR, and the OCR was wrong a few too many times.

AI costs a lot more than all these, combined.

PCs (and screens in general, particularly old CRTs) are a b*tch and a half to film.

You typically film at E.G. 21 FPS, while your PC runs at 50 / 60 (which isn't an integer multiple of 21), so you get Vsyncs somewhere during each frame, making each filmed frame a superposition of two frames onscreen. You also have to be careful about how long the film is exposed for, as, on a CRT screen, the electron gun goes left-to-right, top-to-bottom, and the top-left corner of the frame is no longer visible when the gun reaches bottom-right. If your film isn't exposed for the entire duration of the frame, parts of it may never be visible.

This is why there was a cottage industry of PCs and PC monitors designed to run at Hollywood frame rates. Cathode Ray Dude has an excellent video on why this was a problem and how the problem was solved[1].

[1] https://www.youtube.com/watch?v=qicQUvSUbPM

I think it's partly also about what you see first.

It's far easier to be transphobic / racist / furry-phobic if the first thing you notice about a person is their gender / race / furriness.

If you are that kind of person, you immediately get into the mindset of "what they're saying isn't worth listening to because they're a <slur>, and <slur>s can't be smart by definition." If you first meet them online, you picture them as smart in your head, before you learn that they're black / trans / furry. When you do learn about that, you already think that they're smart, so it's much harder for your brain to dismiss them just because they're something you don't like.

Along with trans people.