Testing is also about looking for things that you don't anticipate.

For me social media used to be a way that I could keep up with what my family, friends, and peers were doing. I saw what they posted and they saw what I posted. Now feeds are overrun with stuff I don't care about (the algorithm) and stuff that isn't real (AI), so it is becoming less and less worth my time to post stuff (my intended audience won't see it, and is also leaving the platforms) and likewise I don't see stuff my peers post. On top of this there is all the political crap and advertising.

This is news for nerds. I am saying that there are viable technical solutions, not that Meta would choose to use them.

The light wouldn't have to be a visible spectrum or one that is interfered with by normal lighting, the camera just needs to be sensitive to it. Even if it used normal spectrum, it wouldn't have to work in all lighting conditions, as it doesn't have to detect tampering immediately, it only needs to detect that tampering has happened at some point in the past. The algorithm could be to wait for testable conditions and then test, if a test hasn't completed in 24 hours, disable the camera and instruct the user to put the glasses somewhere dark, and maybe have a 3 strikes you are out rule before bricking the camera.

They could flicker the LED in some random pattern and then look for corresponding lighting level changes in the images captured. The flicker could be fast enough that it wouldn't appear to flicker to a human.

So we can now non-invasively mess with the pleasure areas of the brain? Has the pleasure (or on the dark side, pain) helmet been invented?

Bitlocker without a PIN is a setup where a normal boot of the machine decrypts the disk without any user interaction. Why would it be a surprise if it was possible to alternate boot from USB and also be able to automatically decrypt the disk? This would be true for any disk encryption software that doesn't require the user to enter a PIN or password at boot time. If you can boot from alternate boot media, pretty much any operating system would allow that alternate boot to have access to the underlying file system, while bypassing the original operating system accounts, logins, and permissions.

So foreign routers and related firmware is too risky to continue to allow for imports, but it's not too risky to allow the same risky sources continue to give us new firmware? Seems to me if a foreign entity wanted to screw us, now with be the time. It's like, those guys over in the US are going to cut us off, so we should use this last chance to send them backdoored firmware.

I have a car that has a rear view mirror that can operate as a real mirror or as a screen to the rear camera at the flick of a switch. I strongly prefer the real mirror because I don't have to shift the focus of my eyes when looking at it. When driving and looking out the front window, my eyes are focused for distance, and with a real mirror I can see what is ahead of me and what is behind me at the same time. The real rear view mirror is part of my situational awareness without effort. However with a screen for a mirror, my eyes have to focus on the screen which is only a foot or so away, so I can't have what is behind me and what is in front of me in focus at the same time. At night the the video based rear view screen is mostly useless as all I can see is washed out headlights - maybe the dynamic range of my video rear view mirror isn't as good as what's in the Polestar, but technically it's pretty hard to match the dynamic range capability of the human eye. I do love my backup camera and it's great for backing up, but for keeping tabs on what is behind me when I am driving forward, I like the real mirror better.

The overall topic of the thread is about comparing the performance of two browsers, and the summary listed electrical power usage as part of the comparison. Your post was discounting the power comparison portion and writing it off as "greenwashing" as if power comparison had no value other than being green. I am saying that power consumption is a valid technical comparison point for reasons beyond the political. It impacts the total performance possible on the machine (my point), and it impacts cost and battery life (points made in other threads here).

I don't understand your comment about the tasks being different between the two browsers. Each browser's task is to render a web page, if one browser uses less energy rendering the same web page as a different browser, it would objectively perform better by this metric. I would hope that anyone comparing browser performance by this (or any other metric), would have the two browsers render the same web page(s), and ensure that the networks were setup so that they hit the same physical target web server and that the target webserver was in the same cache state for each test.

Because power consumption is a proxy for performance? If you run two different programs (performing the same task) on the same hardware, and one uses more watts (watt hours actually) than the other, then the one using more power resources is using a higher percentage of the total capacity of the system. Processors (and their related cooling infrastructure) have a limited amount of heat they can dissipate. While power consumed is not a (speed) performance metric for a computer running a single task, it is a performance metric when looking the capability of the whole system when the computer is doing multiple things. On a fully loaded system, software pulling more watts will bring down the performance of the overall system more than would software pulling fewer watts, assuming the system has thermal throttling, which most modern systems do.

Watched the video and took a quick look at the website and it's not clear to me where the data lives persistently. I get that there is synchronization and it sounds like that there is some logic to keep cached copies of stuff on multiple nodes based on popularity and the numbers of others subscribing. But say I want to host a personal website or some data of mine that I want to share with others. If that data isn't popular enough and I have no subscribers, does that data only exist in my local node, on my desktop machine? If I want to ensure that the data isn't lost, and is available when my desktop computer is off, do I have to ensure that another node is running on a backed up machine in a hosting facility somewhere? I think the answer to these questions is "yes", but that reality doesn't seem clear in the promotional material, which seems to just say "It's on the Freenet network". This isn't necessarily bad, and does have the advantage that I can control who has access to said node, but sounds like I would still be responsible for a bunch of always on and redundant infrastructure, which is one of things that drove people to centralized data models and "cloud hosting" historically.

It checks the "doing something to save the children" checkbox, checks the the privacy checkbox, and check the easily bypassable checkbox. For parents that actually care, they can give their kids devices that have the kid's age properly set (and not give the kid admin on the device), so it provides reasonable path. Sure, some smarter kids could figure out how to bypass, but those kids are gonna bypass anything. For adults, it's a low configuration bar, and doesn't require exposing actual birthdate or anything sensitive to a lot of apps. For lawmakers, they have done "something".