It's the norm because it makes the licensing agreements slightly cheaper.

Well, of course. It was a white couple driving an expensive car.

But, on the other hand, they failed to check that the license plate matched the plate they were looking for. The stolen plate was reported as either "34 DTM" or "34 03 DTM" (the article isn't too clear on this point: the report to Flock was "34 DTM", but it's not clear if the original police report was for that plate or "34 03 DTM"). In any case, these plates did not match the plates of the car that the author was driving.

One would think that checking the plate would be the first thing the police would do before apprehending someone, but apparently not.

So, well behaved, but incompetent. I am not sure that such incompetence qualifies as well-behaved.

If you are OK with Microsoft backing up this information without your explicit permission, then fine, use a Microsoft account. I just pointed out how to avoid this backup (although you will be nagged to set up backups).

I assume that this backup will also include the GUID that was discussed a day or two ago, so even re-installing Windows won't eliminate this tracking.

This is why you use a local login, instead of a Microsoft account.

Microsoft has been making this more and more difficult, but I think that, after setup, one can create a local login and then remove the Microsoft account.

Windows 11 doesn't have an option to turn off updates, only postpone them.

The only way to prevent updates is by disabling the service.

And, what about alternative browsers? Do they send this also?

OK, so what is on the ngrok site that triggers this?

Not explained here, how is this information accessed remotely? Does Edge (or any other browser) send it in a header? Is there Javascript on some websites that makes the target send it?

How do you feel about charging for on-device features, that work without an Internet connection, like this one?

https://www.techdirt.com/2026/...

Except that this feature runs entirely on-device. It works without an Internet connection. They want people to pay to use the hardware they already own outright.

A month or so ago, I received a call about my credit card and, at the point that they wanted to pass the call to their fraud center, I hung up. Obvious scam, right? Well, actually, no. It was a genuine call from my credit card company.

The credit card company has been good with fraud, texting me if there is a suspicious activity and allowing me to text back yes, or no. But then, I tried to buy a new dishwasher: the charge would not go through and there was no text. Eventually, I had to use a card from a different issuer.

But if their own outbound calls appear identical to scam calls, they should be responsible for any resulting fraud.

I could already run Docker Desktop, using WSL for the virtualization layer. What's the advantage of this? Is it only the "without third-party software"?

You might want to consider wiring the money from your non-US bank directly to the recipients. However, even this can be fraught with problems.

Some US banks cannot accept incoming international wires, instead it has to go though an intermediary. In one case, I sent some money to a relative in the US from my UK bank. I tested the setup by sending $100, then sent a larger amount. The $100 that I sent was credited to my relative's account, then a couple of days later (after I had sent the larger amount), they told my relative that money could not be received this way, putting my larger wire into limbo (no, it did not automatically get refunded back to my sending account). What a clusterfuck.

Then, there is the cost of the wire. I have seen that, even if I specify that my sending bank pays all fees, the recipient (my own US bank) may also be charged a fee for receiving a wire.

If international wires can be received directly, they are fast and reliable.

That depends on your definition of "Linux". On my Gentoo desktop [OpenRC + MATE], TigerVNC no longer works, because recent versions require Gnome and Gnome requires SystemD.

There was definitely some bullshit in the arguments. One was that systemd could bring up network interfaces much faster, but when you dig into it, that speed "improvement" was largely due to not running the "arping" command to check for duplicate devices already on the network.