Not that I want to defend AMD but the reason they haven’t been able to is because HDMI is a private spec and the HDMI forum specifically forbade them from releasing the driver they tried to publish a few years ago (even though the firmware with the actual implementation is in the GPU itself and still proprietary). That’s why 2.1 works on windows and not Linux with the past 2 gens of AMD GPUs.

They are quite literally taking the “if you don’t like it, fork it yourself” approach. Who said they aren’t going to make changes/improvements on top of it?

I don’t see anyone else mentioning it but this is also probably because brave browser is published under the MPL license so the licenses are actually compatible between projects. They don’t want to implement completely from scratch because there is a compatible existing implementation that they can build on top of instead of starting from scratch.

Ok but there’s a big difference between being on the bleeding edge of tech and releasing games which are borderline unplayable because of bugs and framerate issues to the point where Nintendo issues an apology on your behalf.

I think I’m cautiously optimistic after seeing the gen 10 trailer, but at the same time after ZA I’m ready to skip out on it if the direction looks bad.

I just want my black/white remakes so I can be done with this franchise -_-

Not that I want to defend Nintendo but Palworld barely even qualifies as the same genre as the mainline Pokemon games, literally the only thing they have in common is that you catch monsters

I was confused about this also, I’ve been using GameNative on a pixel 10 on and off since I got mine in October. Maybe compatibility just got a lot better because of fixing GPU drivers? I did notice a lot of games had issues but I assumed that applied to anything that doesn’t use snapdragon

It depends. I run GrapheneOS and it can pass everything except the most strict integrity check (which is just that you’re using a custom ROM at all).

In practice most apps don’t have any problems. Google assistant doesn’t really work for me but I’ve seen posts saying people have gotten it working. Google wallet and Google Pay are also explicitly blocked by google, so they will never work.

I think the reason GrapheneOS never did a GSI is because most of their security improvements rely on specific hardware calls that GSI abstractions don’t provide access to. This probably would still be an improvement over lineage though, just not as secure as base Graphene is.

The work-life balance is otherwise pretty good and my manager/direct coworkers are chill 🤷

Otherwise I would have lost motivation a long time ago

That’s the thing though. Even if the code is good, the plans are good, the outputs are good, etc, it still devolves into chaos after some time.

If you use AI to generate a bunch of code you then don’t internalize it as if you wrote it. You miss out on reuse patterns and implementation details which are harder to catch in review than they are in implementation. Additionally, you don’t have anyone who knows the code like the back of their hand because (even if supervised) a person didn’t write the code, they just looked over it for correctness, and maybe modified it a little bit.

It’s the same reason why sometimes handwritten notes can be better for learning than typed notes. Yeah one is faster, but the intentionality of slowing down and paying attention to little details goes a long way making code last longer.

There’s maybe something to be said about using LLMs as a sort of sanity check code reviewer to catch minor mistakes before passing it on to a real human for actual review, but I definitely see it as harmful for anything actually “generative”

As someone who has to sift through other people’s LLM code every day at my job I can confirm it has definitely not gotten better in the past three months

NixOS manages to be all of these at once except the manual dependency management

As someone who has worked with a pretty large C# codebase and several smaller ones, I’ve found it to be one of the least efficient languages to program in. This is maybe not a technical fault of the language, but the way Microsoft encourages developing C# means that once you get past a certain point even simple MRs will have 10-20 files changed. There is sooooooooo much boilerplate caused by .NET that even things like Java Spring Boot just don’t have (and even then I’d consider Java to be a pretty bloated language in terms of boilerplate).

That’s ignoring the fact that the ecosystem surrounding .NET is a lot more enterprise-y, meaning a good portion of libraries require paid licenses to use.

My company uses it for some of our legacy on-prem hosting, but a lot of that is being actively decommissioned.

Most of the world can’t feasibly install non-Apple-approved apps on iOS without paying $100 a year so something like that would probably never catch on in iOS land

It’s an Android app patcher that removes ads and adds some other quality of life patches. Primarily for YouTube but it supports several other apps as well.

On YouTube it also adds things like integrated SponsorBlock, extrapolating dislikes, actual resolution buttons, and the option to disable shorts.

What you’re talking about is “source-available.” I.e. being able to read source code but not having licensing rights to redistribute or make changes.

“Open-source” means that being able to modify and distribute changes is built into the license of the code.

For example, Minecraft Java is source-available in that decompiling Java bytecode is trivial - enough so that tools exist which can easily generate a source code dump. However, actually distributing that source code dump is technically illegal and falls under piracy, so it isn’t open source.

Edit: I didn’t see your edit, this comment is kind of pointless, oh well

The code is open source. Nothing is obscured.

“Security-by-obscurity” is a phrase used for any measure that is useless once you know how it works. In this case it’s hoping that a troll doesn’t know about the specific hardcoded rules. None of the rules in PieFed actually work if you are at all aware of them.

Thanks for clarifying, I guess I misremembered the shadowbanning part. I think I was mixing together the fact that reputation isn’t really transparent (users’ reputation can change by even attempting to upload an image that gets flagged, and the vague error means they’ll probably try multiple times without realizing they’re being moderated) and the fact the communities can autoban any user whose global reputation is low enough.

I still think the security-by-obscurity approach to moderation is inherently flawed though, and I hate to imagine how the dev approaches actual account security if that’s their approach to moderation.

Honestly I would consider [user-obscured] hardcoded shadowbanning just as bad.

Just because I’m closer to agreeing with the PieFed dev’s opinions a little bit more doesn’t mean that I’d support shadowbanning someone because the trivially-evaded checks caught a false positive in the crossfire. Piefed’s auto moderation/social scoring is pretty much textbook definition security-by-obscurity. The second anyone knows how it works, it’s useless. It will pretty much exclusively catch people who just wanted to post a harmless meme or something.

At least (for now) Dessalines isn’t hardcoding his tankie beliefs into Lemmy’s source code.

Edit: Blaze is right, it isn’t shadowbanning, but the rest of my point still stands, added the [] part to clarify