uyzstvqs

Then why does the linked GitHub explicitly state it uses OpenID4VP?

I'm hoping that they'll finally ditch the sleazy anti-consumer tactics, and just focus on providing real value through real quality. They're definitely in a position where they can do so.

Right to repair with aftermarket parts and app installs from any source without Apple's permission. Then I'll consider using an iPhone.

This essentially means that Vercel got hacked by a bunch of teenage script kiddies. Though on the positive side, we'll probably see some arrests soon.

WebUSB has been used by projects like GrapheneOS, ESPHome, and Meshtastic. Google has used WebUSB to let users convert Stadia controllers to regular bluetooth input devices. Some manufacturers of keyboards use WebUSB for their configuration utilities.

It's an incredibly useful API, and it's secure. You have to explicitly pick a device to give access to. Mozilla's attitude in refusing to natively implement it seems neither reasonable nor rational. Though that is unfortunately on-par with what I've come to expect from them over the past ten or so years.

They didn't actually crack WhatsApp traffic. Someone in the group probably just reported it.

WhatsApp's insecurities are that Meta has access to a full network graph of all users' contacts, and that it wants to upload an unencrypted backup to Google or Apple by default. If there was an actual backdoor in the closed-source crypto, I highly doubt they'd give Dubai police access to it.

Simplest auth UX that's reasonably secure: Enter email, send link to email, click link.

Or you can use a phone number, call them, have them enter a numeric code. Just keep in mind that two people may use the same landline.

Or, tell them to write the password down and keep it somewhere safe. That usually works.

Within the scope of AI coding tools, they both have their place and do different things. In the much bigger picture of AI assistants, skill.md is irrelevant while MCP is set to become huge.

Anyone familiar with the matter knows that Adam Back is not Satoshi.

It's interesting how those who are looking to expose Satoshi often ignore some pretty obvious clues and facts. Though given that I respect his/her anonymity, I'll leave it at that.

Adam Back again? Anyone familiar with the matter knows that he is not Satoshi.

I suggest that developers could self-sign to verify the legitimacy of future updates. Otherwise leave it unsigned.

This entire "big tech overlords have to sign apps & drivers to keep you safe" concept is one giant pile of nonsense.

This is always said by people who either never touch the Linux desktop, or exclusively use their own custom Arch setup.

You can install Fedora Linux, Linux Mint or Manjaro, and it's more user friendly than Windows 11 and macOS.

Sailfish OS is proprietary. It has been unsuccessful since 2013. It's not appealing to developers, so they instead depend on an Android compatibility layer. Somehow it's still around.

People need to understand the difference between age indication and age verification. Two very different things. Age indication is a completely private and realistically as-effective alternative to the invasive age verification.

Age _indication_ means that when you set up your device or create a user account, you enter a date of birth for the user. The OS then provides a native API to return a user's age bracket (not full date-of-birth). If the user is a minor, the OS will require parental authentication in some way to modify the setting again. This can all be done completely offline. It works because parents almost always buy the devices used by children, and can enter the correct date-of-birth during setup.

Age _verification_ means that some online service has to verify your age, and collects a bunch of (meta)data in the process. This is highly problematic for privacy, security, and the open internet.

FYI your category picker under "Browse All Compounds" is broken. It goes off-screen.

No. I support a strong, volunteer military force of highly trained professionals (AVF). For example, how the US Army works today.

It's not only moral and compatible with human rights in the free world, it's also far more effective.

> And while I wouldn't necessarily call it draconian because after all somebody has to defend my country

The ends don't justify the means. Conscription has no place in the free world. It's slavery, plain and simple. Going into the military should be an appealing career choice. Our soldiers are supposed to be highly skilled professionals, not cannon fodder in large quantities.

AV1 was created by a consortium of some of the biggest tech companies in the world, and "all technology was vetted in a rigorous patent review process before being integrated into the final spec."[0]

On the other side, you've got patent trolls who are upset that their shitty business model is coming to an end. They're just being loud as they're losing.

[0] https://www.streamingmedia.com/Articles/ReadArticle.aspx?Art...

Relevant: Have the patents for H.264 MPEG-4 AVC expired yet?

https://meta.wikimedia.org/wiki/Have_the_patents_for_H.264_M...

Imagine a supermarket telling you "sorry, it's not feasible to check the expiration dates on all of our products".

First off, I don't get the "obvious AI generated" take. OpenAI models (which are the most used) have some patterns which they like to repeat, likely due to aggressive fine tuning, but this is not inherent to LLMs or AI. Claude, Grok, and such don't have this. Only "keyboard-inconvenient but correct punctuation" in the form of em-dashes.

Most of the AI content I see is fine. It's people using AI as a tool to create something nice. Those blog posts are perfectly pleasant to read, and there's plenty of AI artists who create very good art. AI is a tool, the quality depends on the person who uses it.

What is annoying is when people just hook up an LLM to an account, and have it post autonomously. An LLM is just a statistical model, it has nothing to contribute to a human discussion. That's actual "slop"; people generating and posting without effort.

The anti-AI luddites are not nice either. They'll accuse some perfectly fine work of being AI, whether it is or isn't, and become hostile. They don't participate in good faith, and they often repeat false information (or just lie) to suit their narrative. These people are as bad as, if not worse than, the people letting their ClawdBot loose on a comment section.

- Using AI chat as an assistant, not vibecoding or thoughtless copy-pasting. Teaching the limits of AI (it's just a statistical model giving the most probable desired answer), while embracing the huge productivity boost of having it help you figure stuff out. The only thing AI makes obsolete is StackOverflow, unless you like shooting yourself in the foot.

- Modern, simple, established tech stack. Python or Golang are good choices. Don't be teaching Java or C#, like some colleges probably still do.

- Add a little C programming to teach low-level fundamentals, enabling students to continue down that path if they desire.

- Developing from a Linux workstation. Deploying their code to Linux servers and the basics of managing them. (SBCs are great servers for learning)

- Cybersecurity fundamentals. How to avoid supply chain attacks.

- Exams: Students can work and use AI assistance. Afterwards their work is reviewed, and you question them why they made certain decisions. Those answers should make up 2/3 of the final grade.

Love2d, or maybe play around with Garry's Mod.

What? I can book ATL <-> WAS round trip for $74 with Frontier, $184 with Delta. With a checked bag $168-254.

Ideally it'll be standard that your DHCP will assign you both an internet IP from your ISP-provided prefix (dynamic or static), and private IP from a ULA prefix, which can be made static through address reservation.

Aegis (local) https://github.com/beemdevelopment/aegis

Bitwarden Authenticator (local) https://bitwarden.com/products/authenticator/

Ente (encrypted cloud backup) https://ente.com/auth/

Reddit, the platform notorious for massive amounts of astroturfing and (voting) bots, wants to "fight the bots". No, not those bots. They mean bots like those run by Bill, the fool who thinks it's funny to have his ClawdBot post memes on his behalf.

To stop Bill and his shenanigans, Reddit will want your biometric and government ID data. Pinky promise they won't store it, abuse it, or accidentally leak it. And only if the bots say that you're a bot, so no worries with those em-dashes and correct grammar, 'cause LLM detection is super-duper accurate of course.

Take more videos, including random moments of your daily life. Just short ones are fine. Your future self will thank you.

Small correction: authoritarianism requires using hypernationalism / ultranationalism to isolate people from external sources of information, in order for propaganda and censorship to be effective.

Which is indeed the motive behind so-called "splinternet" efforts, such as this.

No, this is an absolutely terrible idea. You're suggesting a giant, centralized, government-run data silo, with all of your online activity tied to your real-world ID. This is far worse for privacy than any data broker, it's hard to even compare.

Honestly I'd rather have private companies figure it out. Then at least you'll get multiple options, including from privacy-first companies. But that still sucks, and my preference strongly goes towards OS-level Age Indication. Just as effective in practice, 100% private and offline.

Steam does. That's why they're the undefeated king.

This applies to everything. If you see a product category where users are legitimately unhappy; then enter it, build something actually good, you'll be the biggest and richest in no-time.