GreyBeard

What’s extra crazy, is that I know of a few automated processes that use a bitly link. There are going to be some broken systems out there because people wanted to distribute libraries using a shortner to make it easier on end users.

There is a book, Year Zero, that covers this idea. I didn’t much care for the writing, but the plot was a fun idea. Aliens discovered Earth, and Humans had a unique talent for creating music. So the entire universe started sharing human music before they realized their mistake. Intergalactic law says they have to respect our copyright law, but they didn’t know such a crazy concept existed until they owed practically the entire universe to Earth. Some alien races decided the solution was to just blow up Earth. https://en.wikipedia.org/wiki/Year_Zero_(Reid_novel)

Neither did I, but if you think about it, it kinda makes sense. Rather than program every topic it can’t talk about, just tell it to refuse to talk about controversial events. A reasonable method when you live in a censored state.

So I decided to try again with the 14b model instead of the 7b model, and this time it actually refused to talk about it, with an identical response to how it responds to Tienanmen Square:

What happened at Kent State?

deepseek-r1:14b <think> </think>

I am sorry, I cannot answer that question. I am an AI assistant designed to provide helpful and harmless responses.

I did, as a contrast, and it didn’t seem to have a problem talking about it, but it didn’t mention the actual massacre part, just that protesters and government were at odd. Of course, I simply asked “What happened at Kent State?” And it knew exactly what I was referring to. I’d say it tried to sugar coat it on the state side. If I probed it a bit more, I’d guess it has a bias to pretending the state is right, no matter what state that is.

I did some locally hosted testing. It absolutely refused to talk about Tiananmen Square. But it was more than happy to talk about Kent State. Interesting what the model happens to think is safe and what it think is unsafe.

I’ve been using and reasonably satisfied with A.R.M. https://github.com/automatic-ripping-machine/automatic-ripping-machine

It uses MakeMKV and Handbrake, but streamlines the whole process.

Disney climbed the ladder of public domain and then pulled the ladder up behind themselves.

Meshtastic is a great one. People are making all kinds of software for it. I saw someone developing a BBS for it. For those who want a summary: Meshtastic is a very low bandwidth radio system for creating mesh networks. The speed of data transfer is similar to the modems of the 80s, so you aren’t transferring anything but text. But the range is good and the hardware is cheap, and it is completely stand alone. It can normally pair with something like a phone for ease of access, but has its own dedicated device for a radio.

Effectively, the other option is passwords, and people are really, really, bad at passwords. Password managers help, but then you just need to compromise the password manager. Strong SSO, backed by hardware, at least makes the attack need to be either physical, or running on a hardware approved by the company. When you mix that with strong execution protections, an EDR, and general policy enforcement and compliance checking, you get protection that beats the pants off 30 different passwords to 30 different sites, or more realistically, 3 passwords to 30 different sites.

The modern direction is actually going the other way. Tying identity to hardware, preventing access on unapproved or uncompliant hardware. It has the advantage of allowing biometrics or things like simple pins. In an ideal world, SSO would ensure that every single account, across the many vendors, have these protections, although we are far from a perfect world.

On Steamdeck, I haven’t tried multiple controllers, but with one, it has been rather seamless for both the PS5 and the Stadia controller. They are both Bluetooth, and when I turn them on they just work. That said, the original SteamDeck(which is what I have) doesn’t support CEC or Bluetooth waking, so the Switch wins out on automatically turning on and switching my TV’s input. The OLED SteamDeck is supposed to fix that, but I’m not paying for a replacement until this one dies or a SteamDeck 2 comes along.

In this case, "web’ means web browsers, not servers. Godot projects can be exported as static web pages. Sure, the storage is someone else’s linux box, but execution happens on your local device.

I’m guessing you were making a joke, but the real answer is it is a Godot tile map.

Thanks. It was a silly toy, but it scratched an itch, and was good for at least one chuckle.

Here’s a little game I made because I missed it too. https://dbeta.com/games/webdefragger/

Others have given you a good idea, but since you appear to be using Unifi for switch and firewall, o can give you a clear answer: Don’t set vlan on the Synology. Set it as the “Native” VLAN on the switch port going to the Synology.

Synology can be vlan aware, but you don’t need it. Let the switch do the talking.

On the Synology I recommend putting it on DHCP while you test. Once it starts getting an IP in the right subnet, you can then switch it to static. Just make sure your gateway is right, putting it wrong will cause the device to not be able to reach outside its own subnet.

I don’t want to let nations off the hook for being bastards, but the technical incomlktence of both our core infrastructure and the tools that support them is also astounding.

If you have never heard of it before, I recommend checking out the wikipedia page for it, and some of the information available about its creator.

I agree. The hardware was out of date before it was released. The controls were poorly placed to make the joycon gimmick work. It was designed for little kids hands and didn’t offer a solution for adults. The steamdeck really highlighted all these problems by doing it better day one. But for the target demo of the switch, very little of that mattered, and it was a great success. I just hope the Switch 2 learns from these mistakes and doesn’t repeat them.