Sounds like a decentralized encrypted messaging platform is needed.

Decentralized probably isn’t desirable for this use case; self-hosted is. When designing something for that purpose based on a decentralized protocol like Matrix, it’s probably desirable to mandate that the most sensitive conversations take place using a server with decentralization disabled and a client restricted to using only that server.

Between Firefox being its usual self and the 11.5gb of VRAM and GTT kwin_wayland is currently using, 32gb does not feel excessive.

Do you need a recommendation for an adblocker?

An app that could be a website and wants a huge intrusive set of permissions? So just like every corporate social media thing ever.

Under other circumstances, I might agree with you, but the guy built a bong into his dashboard.

That’s an impressive level of commitment to DUI.

You’re not wrong, and an open option might be an improvement over the current situation. On the other hand, it might encourage broader use of remote attestation.

I’m mostly disappointed that there’s no meaningful organized opposition. When Microsoft first proposed adding remote attestation to Windows, the New York Times called it out as oppressive. Now it seems like only hardcore open source nerds care, and I think the tech community should be doing better.

I don’t like it. Remote attestation is a violation of the user’s right to control over their own devices. We should be pushing to eliminate it, not expand its use.

Anyone who was publishing to FDroid already is not going to be annoyed about the 24 hour scare screen for users.

Bullshit.

It’s hard enough to get people to step outside the Play Store ecosystem. Any additional friction will greatly reduce the number who do, and the combination of a reboot and a long waiting period is a lot of friction for the average person.

He did it under his real name. His Github username is his real name, with middle initial. He also links from said Github to his .com, which is also his real name. There is no doxxing here, nor is saying I wish someone hadn’t done the thing they did harassment.

I won’t defend the tone of the article though. I find the photoshopped mugshot and name-calling distasteful.

I want him to do nothing.

He doesn’t work for a distribution or a system integrator. He isn’t the maintainer for systemd either. He’s a random contributor, and he works for a cloud company that doesn’t make or sell the sort of devices these laws apply to.

These age verification laws did not require Dylan Taylor to take any actions. He did that all on his own.

I had an inverse experience after an adult beverage or two and talking to someone about a third party’s script they found unsatisfactory. It went about like this:

Zak: filename.py sucks

Claude: What’s wrong with it? Bugs? Code quality? Features?

Zak: yes

Claude rewrote it, claiming it had “multiple issues”. It found and corrected a major bug, added error handling, and improved command line argument handling.

A lot of network, banking, and telephony protocols historically rely on trusting that there are no bad actors in the chain. Technology has added more links to the chain increasing the opportunities for bad actors to tap into it.

Their wish to break the first rule of network security (you can’t trust the client) shouldn’t be everyone else’s problem.

could dramatically cut the energy consumed by artificial intelligence hardware

Decreasing the cost of using a resource almost always results in more use of that resource.

Laboratory tests showed the devices could reliably endure tens of thousands of switching cycles

That’s not very many when GPUs perform trillions of operations per second.

I’ve tried it, and only ran into a couple apps that wouldn’t work with MicroG. I won’t pretend it’s painless, but it’s workable for someone with sufficient motivation.

/e/os is Android without Google proprietary stuff. It runs most Android apps.

A senior engineer obviously needs (and knows how to handle) considerably more access to their workstation and company IT infrastructure than the average employee. On the other hand, I’ve occasionally read complaints from IT security types about engineers being way too eager to install sketchy stuff.

There’s some truth to those complaints. I might need to try out several libraries and tools to see what works best for a certain use case. Is that new one with 15 stars on Github actually safe? Are all of its dependencies? How many developers perform a task like that in a sandbox? How many of those perform a thorough audit before taking it out of the sandbox?

I don’t know if that service can, but LLM-based workflows can do that. Here’s an LLM-based decompiler project which could serve as the first step in such a pipeline.

How much cheaper do you think it should be for not including a 20W power supply? I’d be surprised if Apple’s cost for that part is more than 5€.

because they “care about environment 😉” the €99 charger (which is almost mandatory for a new user) is sold separately.

It’s because they’re required by law to offer it without a power supply. See Article 3a, section 10.

Apple’s first-party power supply isn’t “almost mandatory”, and doesn’t cost 99€. The 20W model shipped with the Macbook Neo in other markets costs 25€ on Apple’s German store, and a generic 8€ power supply from Amazon will work. The power supply most people already have for their phone will usually also work.

It’s changing rapidly, but handing automation tools to people who don’t understand the underlying concepts just gets you a bigger mess. There are no well-established best practices for how to use it safely and effectively because it’s too new and changing too fast.

It will settle down eventually, but a lot of people will do a lot of dumb things first.