There’s still a vocal minority of people upset about all the Linux talk on Lemmy, though I find deadpan is generally a lot less risky here than it was on Reddit.

And respect for keeping the historical record accurate even if it results in an unhappy score. I do the same. 👊

It’s in the uncanny valley. Not the uncanny valley of cheaters vs people parodying cheaters, but the uncanny valley of parodying cheaters to protest the change vs parodying cheaters to say people insisting on Linux are doing so because they do want to cheat.

The negative review is a tip off, but that was my first interpretation of it (that you meant people who wanted it on Linux just want to cheat).

And if you enable the mouse while line numbers are set to be displayed, you can highlight multiple lines without needing to include the line numbers when you paste it later.

I think you’re overfitting to the average here with your expectations. Especially basing that on the experience level of people who would sign up for help learning how to use Windows products. And even then, the ones learning about copy/paste for the first time will likely make more noise about it then those waiting to see if you’ll teach them something new or any that ended up in your training because their work made them or something.

While the majority might lack familiarity, the 40 - 80 age range includes tons of people that have been working with computers (windows or otherwise) since before Windows was even a thing, including many who worked on Windows and/or developed applications for it. Experience will range from not knowing what windows is, knowing it’s the OS but not knowing what an OS is, to understanding what goes on in the kernel at a high level of detail.

There’s a lot of people on Windows just because of inertia and Linux can handle a lot of the use cases. It makes perfect sense to me that someone, once they’ve seen that things aren’t so scary and different on the other side of the fence, would wonder out loud about why they thought their inertia was so strong.

Your skepticism is more baffling to me than that.

Or give a 900 area code.

A compiler making assumptions like that about undefined behaviour sounds just like a bug. Maybe the bug is in the spec rather than the compiler, but I can’t think of any time it would be better to optimize that code out entirely because UB is detected rather than just throwing an error or warning and otherwise ignoring the edge cases where the behaviour might break. It sounds like the worst possible option exactly for the reasons listed in that blog.

I think they meant the other way around, that if you wanted to use it in C/C++, you’d have to either use assembly or some specific SSE construct otherwise the compiler wouldn’t bother.

That probably was the case at one point, but I’d be surprised if it’s still the case. Though maybe that’s part of the reason why the Intel compiler can generate faster code. But I suspect it’s more of a case of better optimization by people who have a better understanding of how it works under the hood, and maybe better utilization of newer instruction set extensions.

SSE has been around for a long time and is present in most (all?) x86 chips these days and I’d be very surprised if gcc and other popular compilers don’t use it effectively today. Some of the other extensions might be different though.

And I had to stop using vscode because of its ridiculous resource usage. I got tired of it filling up my home dir and just went back to vim.

An intern was using it, but I saw that he had set it up to run locally and connect to the ETX we were using and figured he had found a way to avoid that. Nope, turns out it runs a server on the ETX that also likes to fill up the home dir and he also just uses vim now.

Yeah, this has been my experience. I’ve got ABS on my current car but in winter the main advantage they provide is letting me know when it’s time to pump the brakes if I need to come to a stop. The times I haven’t pumped, let’s just say I was lucky that no one else was going through that intersection at the time.

I suspect they are tuned to avoid locking up the wheels in conditions other than ice/slick snow and the pulse is too quick for the tires to get traction when it releases.

This could be intended to settle a disagreement between management people who don’t see the trend of gamers finally getting fed up with the bullshit and others who don’t call the shots but do have a finger on the pulse (or even feel that way themselves and know they aren’t alone).

I’d bet good money there’s plenty of developers and other gamers involved with a bunch of these companies watching decisions being made with horror.

Actually I bet management only allowed this to be a poll because they did notice the trend of gamers getting fed up and previous cash cows running dry, but they needed a poll because they don’t want to believe that the thing they thought was the best way to fight piracy was hated by people who would otherwise be happy to spend money on it.

I always keep thinking back to a piece of software that took weeks to get running at a job where we were development partners and then when I decided I wanted to use it with a personal project at home, I had a pirated copy running within hours. All the DRM stuff just made it into a pain for legit users while those using pirated copies never even saw that after it was cracked.

And denuvo doesn’t even stop sucking once you get it running the first time, it will be wasting CPU cycles and memory bandwidth until the publisher decides it’s not worth paying the license fee for anymore.

I think it would need to be a subscriber service paid for by consumers who are willing to pay for good reviews. Otherwise the consumers become the product and eventually marketers take over.

Also crowd-sourced reviews are what we’re supposed to have already, both on Reddit and Amazon (and anywhere else).

What I envision would be a publication that funds a set of reviewers (maybe a mix of full time and part time, since some products are appropriate for testing as a job while others are more appropriate to just use for a while).

Each product would either be bought by the org directly, or if manufacturers provide review samples, a layer of indirection is used to avoid the reviewer feeling like they need to give a good review to keep the free shit coming (with clear communication to the supplier that free or not well have no effect on the review).

Any issues get included in the review fairly, along with any kind of resolution (which should ideally go through both consumer channels as well as reviewer back channels, the former to show what average customers should expect, the latter to hopefully resolve design flaws).

The reviewer will then keep the product and give updates, either in the form of “still using it and it is like x after y months/years”, “doesn’t get much use because I’m using this other thing instead because of x, y, z”, or “doesn’t get much use because I’m not really part of the target audience”.

My complete vision includes brick and mortar locations where products are available to try out, and maybe sales handled there, where any product available has a “we vouch for the quality of this product” where flaws are highlighted as much as features are.

Though I think the idea is self-defeating because if it gains momentum, it could halt or reverse enshitification and make it redundant, fail, then enshitification returns. Ideally, enshitification is stopped with legislation about quality and enforcement that questions why a bad design is used when a better one is obvious.

Yeah, the only shoes I’ve ever had falling apart (or more accurately, worn until there were holes in one of them) were worn for years before that happened.

I’ve also never spent under $100 on shoes.

And I don’t think it’s smart to buy shoes you haven’t tried on. There’s variation in foot shapes, some shoes just aren’t designed for your foot and need to be “broken in”. I thought all shoes needed to be broken in until one time I got lucky and the second pair I tried fit perfectly right away. Ever since then, I’ll keep trying shoes until I find ones that don’t need to be broken in.

One exception was when I forgot about that when my cousin saw a sale on good sandals and had him pick me up a pair. Was reminded the first time I wore them. I spent a day at an amusement park and my feet were killing me by the end of it. Figured it was because I hadn’t been standing much leading up to that. But then, a few years later I wore the same sandals (now broken in) in a similar situation and my feet didn’t feel nearly as bad.

So try on shoes until you feel ones that feel good right away and your feet will thank you. Spend money up front for quality and your wallet will thank you when those shoes last longer than that amount of cheap ones do.

Also take care of them. If they are tie up shoes, untie them to remove them. If they are difficult to get the heel in, get a shoe horn. If you’re often walking through puddles and/or mud, wear boots. Always wear socks unless your footwear can breathe well.

I’ve never put shoes through the washer, not sure how that would affect the longevity, though it likely depends on the materials.

Good shoes will last longer than the laces, too, so just replace the laces when they get worn down. A new lace colour can also refresh the look.

Being able to and wanting to are different, however.

I think you’re greatly overestimating the number of people who would need to be involved. It could be done by one person in the right RTL design position. ASIC validation doesn’t involve exhaustively searching for any backdoors that bridge between something accessible with low privileges to something that is supposed to require higher privileges.

And if someone else did notice that, there’s a good chance it would just be a “thanks for reporting that, I’ll fix it” without a root cause investigation about how it got there, especially if it gets reported to the one who put it there in the first place.

This argument assumes that they’d only do something if they could get perfect coverage, which isn’t very compelling for me. IMO the question should be “would it give enough access to more information to be worth it”, not “it’s only worth it if it gives access to all information”.

And, as the other commenter mentioned, it is difficult to get some Chinese phones, though not impossible and if this whole line of thought plays into that, the reasoning is probably as much about cutting off their access to this kind of thing as it would be about making it harder to avoid western agencies doing this. They’ve said the first one out loud (they being politicians justifying blocking Huawei), and wouldn’t have said the second part either way.

If it is present there, it doesn’t imply it’s only present there.

And we really have no idea how close of a relationship Google, or any other corp for that matter, has with various intelligence agencies. Same thing with infiltrations by intelligence agencies.

And no, it doesn’t mean that every phone in the world is compromised with this, which wouldn’t be that sophisticated, just stealthy. The sophisticated part would be part of the normal design process, it’s called DFT or design for test if you want to read about it, used legitimately to determine what parts of the chip have manufacturing flaws for chip binning.

Most phones don’t have an unlocked bootloader, and this post is about the data Google is pulling on factory pixels.

Why would they do all the work on the software side and then themselves offer a device that allows you to remove their software entirely? And if it’s worth it just from the “make more money from people who only want unlocked phones”, why isn’t it more common?

Mind you, my next phone might still be a pixel. Even if this stuff is actually there, I wouldn’t expect to be targeted. I can’t help but wonder about it, though, like just how deep does the surveillance or surveillance potential go?

You’re right that it’s pure speculation just based on technical possibilities and I hope you’re right to think it should be dismissed.

But with the way microchip design (it wouldn’t be at the PCB level, it would be hidden inside the SoC) and manufacturing work, I think it’s possible for a small number of people to make this happen, maybe even a single technical actor on the right team. Chips are typically designed with a lot of diagnostic circuitry that could be used to access arbitrary data on the chip, where the only secret part is, say, a bridge from the cell signal to that diagnostic bus. The rest would be designed and validated by teams thinking it’s perfectly normal (and it is, other than leaving an open pathway to it).

Then if you have access to arbitrary registers or memory on the chip, you can use that to write arbitrary firmware for one of the many microprocessors on the SoC (which isn’t just the main CPU cores someone might notice has woken up and is running code that came from nowhere), and then write to its program counter to make it run that code, which can then do whatever that MP is capable of.

I don’t think it would be feasible for mass surveillance, because that would take infrastructure that would require a team that understands what’s going on to build, run, and maintain.

But it could be used for smaller scale surveillance, like targeted at specific individuals.

But yeah, this is just speculation based on what’s technically possible and the only reason I’m giving it serious thought is because I once thought that it was technically possible for apps to listen in on your mic, feed it into a text to speech algorithm, and send it back home, hidden among other normal packets, but they probably aren’t doing it. But then I’d hear so many stories about uncanny ads that pop up about a discussion in the presence of the phone and more recently it came out that FB was doing that. So I wouldn’t put it past them to actually do something like this.

I was just wondering earlier today if Google kept the bootloader open to allow custom OS installation only because they had other hardware on the phone that would send them their information anyways, possibly through covert side channels.

Like they could add listeners for cell signals that pick up data encoded in the lower bits of timestamps attached to packets, which would be very difficult to detect (like I’m having trouble thinking of a way to determine if that’s happening even if you knew to look for it).

Or maybe there’s a sleeper code that can be sent to “wake up” the phone’s secret circuitry and send bulk data when Google decides they want something specific (since encoding in timestamps would be pretty low bandwidth), which would make detection by traffic analysis more difficult, since most of the time it isn’t sending anything at all.

This is just speculation, but I’ve picked up on a pattern of speculating that something is technically possible, assuming there’s no way they’d actually be doing that, and later finding out that it was actually underestimating what they were doing.

When trying to find a copy of Forza 4 (or one of them) after being disappointed with the cut down version they had on gamepass, I discovered it couldn’t be sold anymore because of a deal MS made with Porsche that eventually ran out.

Also set up a standardized licensing process that breaks the mini-monopolies of exclusive content.

Personally, I’d also limit copyright to specific works and not the characters, setting, etc. Then protect trademarks and use those to establish canon. Like in the MCU and DC universes, Spiderman and Batman don’t exist together, but in the Superhero Fan Universe, they are roommates and play genius billionaire vs superhuman with a sixth sense prank wars on each other.