No worries. Yeah, the user interface is difficult. Thanks anyways for the post, it motivated me to self host my own agent, and it works quite well actually.

I find the idea interesting. However I wouldn’t get the commercial service you are advertising. Maybe something that can run on an older CPU as well. Doesn’t need to be fast, but if it uses up the idle compute I have on my rented server, then why not.

A back up on the same drive as the original isn’t a backup. It’s a waste of space, and a stupid idea.

It’s a backup that helps against software errors, but not against drive failures. Depending on what you want to achieve, it’s a perfectly valid strategy.

I have daily backups, weekly, monthly and yearly. All but yearly is deleted after a certain time, but yearly I keep forever.

They usually do transcoding when you watch a movie or so. Depending on hardware that may heat it up a bit. I guess if it would have heating problems then OP would have resolved that already though.

Proxmox is a commercial product.

And they know you are watching them?

Now that’s even more stupid than calling it “AI”

What does S stand for?

I’m personally all for rebranding “AI”, because current “AI” is not intelligent. Before current “AI”, the term AI meant that a computer would basically have human skills combined with computer skills, and current “AI” certainly falls short in many human skills.

It’s a nice survey, but often forgets that you can also self-host in the cloud, and doesn’t even consider that rented root-servers exist.

AI: “investing in AI was a mistake, invest in green and left things instead!”

Investors: “Nyaaaaahhh” throws data centre into fire

I think you may have too high expectations about the motivation and work ethics of the average office worker. Just because they’re job is too decide who to kill does not mean they are extra careful about that. Killing is part of the job after all.

I kind of believe them because it seems quite reasonable that the AI or whatever draws a slighty wrong border around a military complex, or an AI that was trained to distinguish different military targets was not trained very well on schools.

At least they are investigating. I assume that while the military naturally accepts that some civilian deaths cannot be prevented, they still try to minimise those when possible.

Since people all over the world are prone to simply trusting AI, it is important that a military creates procedures and educated their responsible personnel in order to mitigate the risks of AI use.

That’s a very important point to consider. Unless you remove the battery, you shouldn’t use it as a server. If you still do, store it in a metal bowl or so and put it somewhere where the heat of the bowl itself cannot start a fire. And make sure that your smoke detectors are still working and not expired. Then, be aware that if the battery explodes, your apartment fills with biting toxic smoke within minutes, and that can happen in the middle of the night. So keep some clothes ready at the door that you can wear when fleeing the apartment.

Letsencrypt can be directly forced to revoke certificates by the US. Organisations outside of the US are less vulnerable against that.

So then there’s gonna be an option for a strictly chronological timeline?

I’m using colota, but it’s still very noisy.

I’m currently using reitti, which simply does not deal with the position reported by my phone being noisy. I get all the noise shown on the map. Does Dawarich do any denoising?

Keep the slop, kill the science. Got it.

Arguably, approaches to optimisation problems such as AlphaFold are inherently biased and not really suited for real science. While other approaches take longer, their biases can at least be described. It doesn’t help if AlphaFold is super fast, but then you don’t really know if you can trust it’s output. It’s just like ChatGPT which is often correct, but that doesn’t help if you never know when it is correct. However, with protein folding at least you can compare solutions and easily see which is better, because computing the potential energy of an existing solution is relatively simple. The hard part is to find a good solution.

Anyways, AlphaFold surely has its uses, and surely works well for proteins that fold in some sort of “standard” way, similar to the proteins in the training data set.

I see, thanks :)