

Now you’re going out of your way, to find more dubious third party scripts, instead of going through the official docs, just to try and prove a point wrong which I haven’t even made? Welcome to my block list.


Now you’re going out of your way, to find more dubious third party scripts, instead of going through the official docs, just to try and prove a point wrong which I haven’t even made? Welcome to my block list.


Yeah I don’t think there is anything of value to be found here. I added another comment highlighting the problems I see with this test here.


Adding to this, as I have now skimmed over the original video in question by ANA[DIA]LOG: He already introduces the topic by clearly stating that he is strongly biased towards “cables make a difference”, I can only assume he lacks the necessary fundamentals about what a cable does. He describes his test track to have “a lot of frequencies”. We don’t know about the gear he used, and he chose a singular test which cant’ be replicated as it has too many muddying variables. The data he shared had to be sanitized by you, to even give any chance of an objective interpretation, which really is impossible in these circumstances. When his test shows some differences, rather than being surprised and questioning the sanity of his test-setup, he just accepts the result that satisfies his bias. This is not science, this is a grifter trying to sell snake oil to the gullible.


Maybe you’ve heard about a page called Audio Science Review. The owner amirm did some tests with a much more sane setup which he describes in detail. He actually finds his high priced cable (well above 2000 USD in this case) to perform worse than the cheap ones if anything, picking up more noise. I will trust his objective measurements any time of the day, this guy actually knows what he’s doing.


Blog post about this event by the someone in question: https://theshamblog.com/an-ai-agent-published-a-hit-piece-on-me/
This is highly dystopian. The article makes a very important point: It does not matter that AI is just imitating humanity. Their output has real and possibly destructive consequences for us.
I’m beginning to view the current internet as a whole to be lost. We’re already starting to put up walls so that these models can’t get in anymore, and precautions are only going to become more sophisticated as AI keeps finding and abusing holes. Let’s hope that this is the return of a small and personal internet.


You’re not the author and I still think you should not have posted this, especially not in this community.
The post makes heavy use of Cloudflare and Docker, which are both either unnecessary for the task or have more open alternatives. The post mentions AWS, Google Cloud, and Azure, which is fair, and was not a point of my criticism. Using a European server hoster just doesn’t even meet my idea of a bare minimum for a community like this.
You want a European stack? Use Proxmox, which is German, or just straight LXC. Cloudflare really serves no purpose here, kick it out. Then host your sources on Codeberg which is also German. How difficult was that?


Then disable password auth and ban their IP? SSH is not even a common intrusion path, those crawlers look for open gates, nothing else.


I don’t like this. The post tells you to do so many unnecessary things and gives zero explanation. Of course AI is mentioned in the first paragraph already, and this screams vibe coding with no understanding.
Why the Cloudflare stuff?
What do you need request caching for?
Why use a nonstandard SSH port? Obscurity is not security.
What is the purpose of all the other SSH config changes, which heavily alter the defaults in a way that people might not expect?
Your configuration makes the system reboot automatically, but you fail to mention that.
Then the rest just quickly glosses over setting up some obscure container management frontend and deploying something to it. Again, no explanations. Who needs this?
Edit: I only now realize that we’re in /c/buyeuropean. Why would you recommend Cloudflare, Docker and GitHub here? Is this a joke?


It does have a lot of improvements but sadly literally falls flat on what made the first one so enjoyable to me. I loved having to think about my routes and using tools in order to traverse a train that was not made for traversal. Now everything basically has a more or less direct walking path. You get the truck very early on and it will be your do-it-all almost the whole way through. I bought a PS5 just to play this game early and I haven’t even finished it.


Don’t use public trackers is really the most important precaution imo.
Windows dwm. I’m not using a DE within WSL.
Mono block units are, yeah. However some split units have appeared on the market in recent years and they are great for people who rent an apartment without any AC. I bought a Midea PortaSplit last year and it’s amazing.
Okay but this is really just a bugfix release. What’s the big news here?
I could probably make something work, but I don’t see a reason for it at the moment. My current setup works pretty well. It would be nice if Jetbrains started fixing stuff instead of just throwing new broken features onto the pile but that’s just not gonna happen.
I’m using IntelliJ from within WSL, and while the Wayland mode seems to work well and look better than X11, unfortunately I can’t snap the window using Fancy Zones. That seems to be a WSL problem though.
Someone probably will try to recommend the WSL integration in Windows-based IntelliJ. I can tell you it’s still very buggy and makes the projects I’m working on break.


Ja, aber pünktlich zum Renteneintritt Krebs kriegen ist nun auch nicht, wie man sich den Ruhestand vorstellt.
No it’s not. That’s what
/var/log/auth.logis for.