A quick web search suggests the native Art Mode function is probably the best.

Ideas are cheap. “(site) but federated” isn’t novel, and in a lot of cases you don’t really even explain what that looks like, and I can’t make it make sense either.

We all live busy lives. I wrote the last comment on the bus, I’m writing this one from the toilet. If you want to contribute to the fediverse, great! Set aside time and contribute to a project, either by extending existing ones or starting your own. There are plenty of minor bugs that need fixing, and if you are proficient in any languages, translations are always needed, no coding skill required.

What is the point of repeatedly making these posts, when you have no intention of actually doing any work to implement them, and hardly even engage with people who reply? I feel like all you’re doing is sapping the time from people who do engage.

Removed by mod

I got one so I picked it apart. The link runs a script on the local system that downloads a kubernetes runtime. From there I assume it runs a Bitcoin miner or something, I didn’t go any further.

Pretty much all of them.

I didn’t know haproxy had a GUI, but the config files are very simple. I would just modify an example one and call it good.

I’ve only seen it for two so far, and after I blocked them it was fine.

Part of the problem was Sonarr trying to grab episodes that hadn’t aired yet. I think a fix for that is forthcoming. That should cut down on them. I don’t know if there is a fix to let you block extensions.

It was a little different, because the quests were procedurally generated. Unfortunately most of them were pretty boring, being either a “fetch the item” or “kill the bandits” quest. I don’t know another major game that did that.

Yeah it doesn’t look at file extensions. Block the release group name. Those releases are malware.

I have a Jonsbo N3 and I’ve been happy with it so far. The N1 and N2 only fit four drives, so if you want more, you’ll have to get the N3 as well. And you’ll have to get an HBA or something, because your motherboard will probably only have four SATA ports.

The biggest power draw will be the drives, also. And you don’t really need ECC on the desktop; random bit flips are uncommon and rarely significant when they happen. Your filesystem and/or RAID should protect against disk corruption.

Interactiviry seems to be a good thing. What brings you to participate here on Lemmy?

For http(s), yes. Other services that don’t support host routing, which is most of them, no.

The thing you’re missing is runtime. If you match your usage to the UPS exactly, your runtime will be zero. If you consume 300W, your UPS needs to provide 300W for however long it is until the power is back. All major UPS companies (APC, Eaton, Cyberpower) have calculators on their site.

But a UPS for extended runtime will be very expensive, and lead-acid batteries must be replaced every five years or less, else they will not be able to sustain a load, so budget for that maintenance too.

A UPS is really just to bridge the gap between mains power going out and a generator coming online. A large-capacity UPS will be prohibitively expensive.

Also, note that cited power figures on the NUCs are probably peak draw. They will probably draw less on regular usage; you’d have to measure it to know for sure in your specific use case.

This would be better implemented as a function in Lemmy itself. A bot shouldn’t have the level of access necessary to do this.

By signaling to oncoming traffic and vehicles approaching from the side, a front brake light provides an essential visual cue that a car is slowing down or preparing to stop. When the light is extinguished, it indicates that a stationary vehicle might initiate movement. According to Tomasch, this visual feedback can significantly truncate the reaction time for other road users, leading to shorter stopping distances and consequently diminishing the likelihood of accidents.

Sounds reasonable. Personally I just want front turn signals to be visible from the opposite side again.

For a limited number of requests, yes.

So who decides which facts should be included in the training data?

You can tell a lot of lies with only facts.

That’s not possible. Any model is only as good as the data it’s trained on.