The only thing I didn't like about systemd was the unfamiliarity of the commands. Typing systemctl and a verb and a service and some more arguments feels a bit clunky.

But otherwise it all seems to make sense to me - lots of little commands running with least privilege, very simple description of services in unitary files, secure binary logging with proper filternig, proper utilisation of cgroups to reliably kill processes and enforce limits, parallel startup/shutdown etc.

Most of the objections I've see to it online seem miss the point and ignore answers which would solve or refute their objections.

It's like the Amish but with operating systems. Technology has advanced this far but must advance no further.

Systemd and Wayland have become ragebait for some people.

The purpose of a duress code is plausible deniability.

There is an encryption product for Windows called VeraCrypt (aka TrueCrypt). You can create a hidden decoy volume inside an outer encrypted volume which has all your stuff in it. When mounted an observer cannot tell the difference between the hidden volume and the real volume since they are mounted the same way with different passcodes. You can put files in the decoy for plausibility but not the things you actually want to hide.

The same should be true of a duress code in a phone. It should be possible to put files, apps and stuff in the decoy that show activity e.g. email, pictures etc. Providing the other profile is wiped while this screen is showing then there is no immediate way of proving it was a duress code.

It would have to at least convince the border guard, but it should withstand forensic analysis too. So it might be necessary to do what VeraCrypt does.

Most likely option is it flops out but only if the angle is sufficient for the cube to overcome its centre of gravity and topple. And do so before the orange portal rises again. Also depends if the downward orange portal directly touches the base or not, since there might be a bit of a "lip" to overcome too.

I really want one or more of these AI companies to fail hard. I think their only strategy is literally to burn through billions hoping it'll be another company that goes to the wall first and they'll somehow reap the rewards by hanging on longer.

If the duress pin makes obvious it's the duress pin then it's not really doing its job. It should instead open a profile with not much in it while erasing the other profile and files in the background.

That's a pretty lame contest. Contestants were told to make notes from an exact list of famous Europeans or birds with some lame architecture on the other side. Can you imagine the excruciating process of finding and sorting either list - "Marie Curie can't be there because she left a negative comment around the cleanliness of her room at a Belgian conference!", "Our prime minister died from an allergic reaction to curlews!"

it's very important to update your AI context, so that the next time you crank the handle it can issue platitudes that resemble relief upon your safe trip.

Reminds me of those ads where scammer pretends if you invest in some obvious scam you too will own a jet and a Ferarri just like him. If it was such an amazing way to make money the last thing they'd do is tell anyone else about it.

When I cook, anybody else in the kitchen who is there not there by invitation is just an annoyance. So people can be there if asked, otherwise FOAD.

It's not just that people are "perverts", it's that they're wearing a camera on their face constantly filming people who have a reasonable expectation to privacy. Even in public most countries would protect that expectation in a lot of cases.

So unless somebody wants to be violently assaulted and their glasses ripped and smashed off their face, then it might be best to not buy them at all, or only wear them in private.