What’s wrong with lemmy.ml? It’s a pretty generalist instance if you ask me. The only issue I have with it is that it doesn’t block obvious troll instances like lemmygrad or the one that’s even worse by default but you can do that yourself these days.

If anyone reading has proof of M$ spying on the German government they could whistle about, right about now would be a great time to do it ;)

Hell seems to be freezing over at an alarming rate these days; climate change is getting pretty extreme down there too huh?

More likely is the device firmware and you likely can’t fix that.

If you have a reasonably up to date mesa and use a Proton version with a new enough DXVK, DXVK can utilise Graphics Pipeline Libraries to link shaders just like a d3d11 driver on Windows would, eliminating stutter.

I believe shader precomp is used for some video codec edge cases though, so YMMV depending on the game.

No, it wouldn’t make any sort of difference.

And also in any other filesystem’s code or the block layers below the filesystem. As I said, unlikely scenario.

Also, their client is still open

*is open again. The clients they distributed were not open source until they open sourced sdk-internal. The fact that you couldn’t even build it with only open code even if you wanted to was a bug but that’s a rather minor issue in comparison.

I also fully believe that they would not have GPL’d sdk-intenral without public pressure. Even when they were originally called out they were pretty clear that the integration of proprietary code was intentional and done with the knowledge that it would typically violate the GPL.

If you don’t see what’s ethically wrong with even attempting to subvert the GPL, I don’t think you’ve understood open source.

Until the situation now, this was limited to the server, not the clients. You could replace the server with Vaultwarden and build it without enterprise features. Not ideal but fine because the server isn’t the critical part. It never handles your secrets in any way.

What they tried to do now was integrate proprietary code into the clients that everyone uses. This is a lot more critical as it can access the secrets in plain text.

This also wasn’t a “mistake” or “bug”, they openly admitted to doing this with the intention of subverting the client code’s GPL.

One does not “accidentally” build a proprietary SDK for months and make the clients depend on it, intentionally violating the GPL.

They even publicly admitted to doing precisely that, defending their GPL violation with dubious claims how the GPL supposedly works.

For ~$30 a month, that’s a complete and utter rip-off.

Even here in Neuland Germany you get at least decent internet with no caps for that price.

There aren’t any “extra access checks” to my knowledge. It’s just the same regular access checks applied to a different set of circumstances.

Flatpaks are containers. They do have a lot of holes though.

As long as the hardware functions as it should (e.g. respects barriers) and there is no software bug in the stack, no.

That’s a highly unlikely scenario though. Make backups.

A driver manager will not make the problems inherent to Nvidia’s crappy proprietary drivers that need workarounds go away.

If you don’t want to tinker a whole lot, buy a GPU from a vendor that hasn’t been actively hostile to its users for decades and is well supported by Linux and the freedesktop such as AMD.

No AMD GPU user has a need for anything resembling a “driver manager”.

Please stop trying to interpret the SMART data report. Even if you’re knowledgeable it can easily mislead you because this is vendor-specific data that follows no standard and is frequently misinterpreted by even the program displaying the data.

If the self-test passed, it’s likely the cable or the controller. Try a different cable.

If you want fast nix evals and docker builds, you absolutely do care about per-core performance.

I have little experience with Rust but, while I do know that it parallelises quite well, I also believe that there are still many single-threaded operations on the critical path though such as linking.

I think for your purposes, you’ll do well with any 8-core AMD Zen4 that can draw more than 45W or any 4+n core Zen5. The latter would be a bit more efficient but practically the same perf in code compilation.

Intel is not competitive currently and ARM is still not quite there yet.

They meant the SMART self-test, not SMART data readout. Those are not meant to be interpreted by laymen and often not even experts.

What are you going to do with it that requires multicore perf?

as an independent voter that feels continually ignored by the by the right and left

A party in the U.S. of any relevance that could be described as “left-wing” would be news to me.

You’ve got a corrupt conservative party and an extremely corrupt "pro"gressive(regressive?) anti-democratic party.

third parties can be an attractive choice for some

Third parties are never an attractive choice for anyone in a first-past-the-post voting systems with two extremely dominant parties, regardless of what any of those parties stand for. The only sensible choice is the (in your opinion) least bad option that still has a realistic chance of winning.