Your ad hominem attacks aside, The point the author is trying to make is that Microsoft is actively using dark patterns in an operating system to get them to buy things.
If you simply disable OneDrive without correctly uninstalling, the system will blast notifications at you with an ominous warning "You could lose data your system isn't backed up!"
PowerShell and most CLIs are terrifying to non-technical people. Literally Here Be Dragons. The layperson might be skeptical of a YouTuber telling them to run a dodgy script, but in the age of "delete system32" people sure as hell aren't going to run a command as admin that a user on a random forum recommends they run.
Stuff like this is why I have moved all of my systems except my gaming PC to Linux.
RE "....If you simply disable OneDrive without correctly uninstalling, the system will blast notifications at you with an ominous warning "You could lose data your system isn't backed up!"....."
The implication here is, OneDrive is backup . It is NOT. Moe Microsoft dodgyness ....
The publisher for this game was Activision. They absolutely had deadlines, lots of (1987) money invested in this, outsourced to a third party company in Hungary, had the outsource team fail, moved development platforms a few times, wrote a programming language and a game engine, and then became the best selling C64 game.
When Windows 11 was force-installed on my main game development desktop, I was skeptical, but kept using it. I was annoyed at having to turn off all the tracking and noise (like news articles)
When it updated and started shoving AI down my throat, with no easy way to turn it off and suddenly lots of data I don't consent to sharing getting used, 11 became the last Windows OS I'll ever use.
Whenever the next version comes out, Im moving fully to *buntu.
My main laptop already uses it and Steam on Linux has been fantastic. Any bugs or issues Ive experience have been due to my very unusual setup (like an eGPU over Thunderbolt)
On my wife's laptop, I've uninstalled MS AI 3 times. I'm just about to lose my mind. I'd have already wiped the machine and moved it to mint but the data in her one drive, bookmarks, etc, I'm sure migrating her over won't be a totally seamless experience. I also have not tested battlenet under linux wine in a long time, and I expect some level of anti-cheat to give me hell there.
On all of my machines bar one, windows is completely gone. I have a simrig, currently running win10, but the hardware there, simucube base, simucube pedals, require some drivers I don't believe exist under linux, and/or don't work properly, and then there is iracing with it's easy anti cheat usage, from my understanding I'm screwed there as well. So it'll live on Windows 10 until the day iRacing stops supporting windows 10, or start supporting linux.
after having written that, I wonder if the simucube tools will just work under linux anyways, the UI is all written in QT, maybe simucube has/is developing linux drivers, given they're finland based :) .. I'll need to test it out
Simucube uses the hid-pidff driver which is built into the kernel. For setting up the base using the SC2 software there is a guide available[0]. I’m not an SC owner myself but there are a few people on the SimRacingOnLinux[1] discord who seem to have everything working nicely.
You can install programs under Steam that are not distributed through Steam.
You can install battlenet under Steam and use all the proton magic to make it work. Starcraft 2 and diablo 2 both work very well (those are the only two I've tried). At least for SC2, anti-cheat did not cause any issues if it's even there at all.
Seriously the only thing stopping me from putting linux on my wife's laptop is the fact that she uses a cricut, which has software that doesn't work well on linux.
Also, I really dislike cricut as a company. Such a scammy business model.
AC works fine just requires a little extra setup, either use this script[0] or use the latest GE-Proton (with a fresh prefix), I recently updated protonfixes to fix a CM/CSP issue. The latter is better as newer Proton has some definite performance improvements.
Fresh should mean creating a new prefix. For example, Proton-GE enabled wow64 in Wine by default, but it requires recreating the prefix to use it. Should be easy enough with Protontricks or even Winetricks.
Yes very much this. It is possible to modify an existing prefix but there are quite a few things to do, it's easier to back up savegames and game config files and then create it anew.
Thank you very much to both of you. Got it running flawlessly with CM on Steam Deck OLED at 90fps using GE-Proton10.34 and LSFG-VK for frame-gen. This one goes straight to the top (alongside DR2.0 and AC Rally) as favourite gaming on the go.
In my (admittedly limited) experience, it does run, even with quite comparable performance, but getting a wheel to interact with the game has been a bit challenging. But this could be resolved with a custom driver for my specific hardware.
Using the community standard mod manager seems to resolve the UI jank by completely bypassing it.
yeah, I've heard this too.. and I'd rather my rig just works rather than try and stuff around making it work under linux + I know iracing is cooked anyway, and I've spent enough money on the rig to just want it to work, and not get stuffed over by some anti-cheat, maybe soon
That matches my experience almost exactly. I was hanging onto Windows almost entirely due to cutting edge graphics and my Nvidia card on my desktop that I'd built.
Windows 10 was already pretty bad, but it felt fast and stable. I think they started putting content in the start menu, and I think I did regedit stuff I can no longer remember to get rid of it.
Windows 11 they made us upgrade with a gun to the back of our heads, they made it feel sluggish, they hid settings in such a way that you're expected to use Search to find the setting (although Apple has that issue too), and somehow the Search wants to include the whole Internet instead of what's local.
But the AI agentic force-feeding was the last straw. What am I, at work?
And then HN insisted Linux gaming was ready and they were right! Someone wrote to me in a comment, "join us, brother" and I'm glad I did, it's brought joy back to using my machine and playing around again.
They already crossed your line with 11, and you're still using it despite Win10 or Ubuntu also being an option. Are you really going to switch when 12 comes out, or is something holding you back?
Ubuntu on servers has always been "a choice", Debian is definitely the preferable of the two. Even on desktops, I'd sooner suggest Debian or Mint than Ubuntu. Ubuntu is a dead distro coasting on a reputation 15(+) years out of date.
(And it used to be that Ubuntu was still a defensible choice for maximizing the chance of getting help online, but LLMs have effectively neutralized this advantage.)
Mint still uses Xorg, so it's outdated. I tried it recently, it wasn't working with my iGPU+dGPU (nothing exotic, just a regular PC), and all the other distros already went to Wayland so nobody was talking about this online. I feel bad for anyone who gets convinced to switch from Windows to Mint, being told it's the easy one. The fix was to just install Ubuntu.
Maybe Xorg is inherently better than Wayland, but that doesn't matter, the ship has sailed and the community evidently doesn't have time to properly support both.
I genuinely don't think Xorg is a deal breaker for newbs and I would characterize dual-GPU as at least slightly exotic, maybe because I've never owned such a computer, but that's a fair enough point. Personally I think the polish of Cinnamon makes it the best recommendation for somebody new, and I know a whole lot of people start with that and have a sufficiently good experience that they stick with linux (while maybe moving on to other distros.)
It's not exactly dual GPU, just the Intel CPU has integrated graphics as usual. I'm not surprised if you don't have that, but it has to be the most common desktop setup, and quite common on high-end laptops. Was giving black screen after wake. Probably a solution exists somewhere, but even if I found it, the fact that this was broken out of the box and didn't have a clear fix was already reason enough not to trust it.
The GUI layout of Cinnamon vs KDE vs w/e seems like the main thing people argue about, but it doesn't matter compared to this. Anyone who even knows what an OS is enough to go install Linux will figure out how to use whatever GUI you give them, provided it works. The bar needs to be at making sure stuff isn't straight up broken.
To be honest I haven't owned a dGPU in almost 20 years, but I've been lead to understand that most users with them use them all the time and ignore their iGPU, unless they're laptop users, in which case they might have to use Nvidia's proprietary drivers from what I understand; the installation of which is something Mint makes straight forward for novices, or so I've been lead to understand. Maybe I'm wrong about some of that.
I definitely agree that KDE vs Cinnamon probably doesn't matter. But I'm afraid I don't think particularly highly of any KDE-first distro; it's great from, for example, OpenSUSE, but that's not a distro I'd recommend to new users for other reasons.
The problem I've got with Ubuntu is they keep doing weird shit like submitting desktop searches to Amazon or putting ads in the motd. They're an erratic organization and I think it's a mistake to send new users in their direction. Mint may not be perfect, but I think it's broadly inoffensive and mild, a good distro to leave a good first impression on a new user fumbling through the process themself.
Imo Ubuntu deserved to lose its users when they switched to Unity, not because Unity sucks (it does) but because it's unacceptable for a newbie-focused OS to rug-pull its entire GUI like that for any reason. But it's still #1, so realistically the leader is going be either Ubuntu or something corp-supported like SteamOS.
I don't think this is a problem at all. I tend to install Debian from the command line (Arch-style), but from what I remember GNOME is the default DE. DEs are largely a matter of opinion, but I find GNOME to be more polished overall. I do use a few extensions however to recreate a desktop-centric UX (Dock, boot to desktop and a few other tweaks).
Debian is a soft no, because despite being an excellent distro, it defaults to GNOME or the user has to deliberately choose something else, which is a problem for giving distro recommendations to noobs because whe you start tacking on stuff like "and make sure you enable the..." their eyes start to glaze over and you risk them thinking the whole affair sounds more complicated than it really is.
I mean I really do love Debian, if not OpenSUSE I would be using Debian now, but it's not a great distro to suggest for absolute novices.
He's not wrong though, the amount of Snap stuff you have to remove in a fresh install is starting to get a bit annoying (I usually remove at least the Snap versions of Firefox and Thunderbird and replace them with binaries from Mozilla - they will still self-update).
You are right, the snap versions mostly work fine. It's just that there are a lot of annoyances due to the nature of Snap packages (slowness, increased disk space requirements, problematic integration with the rest of the system...), but it definitely is possible to live with them.
My Gentoo system is fully systemd and Wayland based from the start. Might sound like heresy to some users, but it was my decision from the start as I liked how they worked, that they are the future, and that you don’t have to wrangle shell scripts for building an OS. I had used systemd a lot via many Ubuntu servers before, so that helps.
> Canonical announced that they are no longer using Debian as a base
When was that? I don't disagree that it appears to be the case (especially with replacing coreutils/sudo/etc and the... varied approach to .deb vs snaps) but I'm not aware of them saying it explicitly in those terms?
Is your name a reference to the Blizzard game? If so, I worked on that :)
You're not wrong, but tbh I'd move upstream to Debian. I use Termux on my phone (Z Fold) with Debian and XFCE, and have been extremely pleased with the performance. Combined with a folding keyboard and some AirNeo's, it's become a fantastic micro-development system that fits in a hand bag.
Not that I don't like Arch, it has a very few (subtle!) things that Ubuntu has solved recently, like eGPU hotplugging
Sorry but this comment is part of the reason anyone should rightfully be scared to switch to Linux. Not only do you have to pick one of 999 distros, but every choice is wrong according to someone. Which one do you recommend, and is it the kind that will throw random issues or be called evil?
These are good choices, also consider Arch if you want the most agency over what goes into your system. That being said, you can also build a very minimal system with Debian from the command line with the arch-install-scripts. It's just that Debian stable will freeze relatively old packages for the sake of avoiding breaking updates, such as changes in configuration files that require manual intervention. On a gaming rig however, you will typically want to avoid Debian as you want the latest drivers, latest Proton/Wine, etc. as the performance uplift can be substantial and compatibility keeps improving.
For the most agency over my system I prefer Qubes OS. I use Debian and Fedora inside VMs. Their mnimized versions are available from the Qubes repositories.
I have had an excellent experience with KDE on Fedora. Has been stable despite being on the forefront of updates, familiar UI approach for Windows refugees while still offering plenty of customisation options for those who seek it.
If you have any unusual set-up going on personally I'd recommend a rolling release distro like manjaro (arch) or fedora, so you get latest drivers and whatnot fast. Modern releases of these distros come bundled with the same desktop environment options as Ubuntu and good, easy to use package install and update GUIs. IMO it's more noob friendly than Ubuntu because your stuff is more likely to work without weird workarounds.
Fighting off snaps would be reason enough to abandon them but Canonical has control of the snap store in a way that is antithesis to open source as they're trying to run a walled garden play. This is the exact type of crap that lit a fire under my ass to get off Windows in the first place.
It's hit-or-miss, with recent live service and esport titles being the iffiest. The older multiplayer titles, casual shooters and natively ported games are more consistently supported and form a sizable library of working online games.
i thought it would bother me, but honestly, tehre are just too many good games that dont require eac.
i would imagine eac on linux will have to be addressed once steam machines drop, but for now i look at it like, if a game requires eac, at this point the game studio is just too lazy or cheap [0] to be linux compatible so we just play something else. far too many great games.
[0] its even more silly considering eac doesnt seem to stop cheaters at all. every single popular game that requires eac is still absolutely overflowing with more than obvious cheaters.
It's not the same as EasyAntiCheat and doesn't support the same features. It's like saying Excel works on iPad, but you can't even use VBA on that.
Or a game example: I have Minecraft (Bedrock) on my phone so therefore I should be able to do the same things as Minecraft (Java) on Windows. The problem is they're the same names for different software with similar, but not the same, functionality.
So you're saying that easy anti cheat on linux is different from on windows? I am aware it is not as effective as detecting cheating on linux, but does this affect gameplay itself? Or do game developers not want reduced efficacy of detecting cheaters, and so they don't support linux at all?
I don't play those games myself but the word is that the EAC on linux lacks the same kernel hooks that are available on Windows. I personally consider that a plus but if you're a developer obsessed with strong anti-cheat you probably do not.
Linux kernel provides ways to observe from user space. The problem is that there’s nothing to stop someone running a kernel which neuters anticheat tools ability to observe using that functionality. As far as I’m aware the only way to mitigate that is via measured boot attestation and having signed kernel etc.
The bigger problem isn't whether or not it can be traced, or if it should be traced. It can and it should. The central issue at stake here is the sale of government secrets to private and (presently) anonymous groups for profit. The author is stating that since we don't know who the trader of that commodity is, and because that commodity's price is tightly coupled to actions related to the war, that trader could be helping enemies of the U.S. Without the knowledge of who the person is, or how they knew to make such a huge market movement, a claim of treason can be argued.
The biggest problem is the intelligence community is heavily rooted in trust. Movements like these signal there's an intelligence leak to the general public, or more appropriately, someone with $580 million lying around. An intelligence leak reduces trust; allies are less likely to share information if it's leaked. Conversely, trust is returned when the leaks are found and plugged, and measures put in place to prevent those leaks in the future. The author is stating that these leaks are unlikely to be plugged, which will reduce trust in American intelligence. After all, as the president said, "Let's say I was gonna do it or let's say I wasn't gonna do it, why would I tell you?" (https://www.bbc.com/news/live/c4gqjyk0vx3t)
Except he is telling someone, and that someone is making a lot of money.
1. Brokerages, banking, and the financial sector in the U.S. is tightly regulated. Personal financial information is very private, and financial data breaches are treated seriously (See: Equifax). You shouldn't be able to publicly trace it. It won't be disclosed unless the owner of the account consents (like a credit check), or a subpoena is issued. If the trader isn't suspected of a crime, there's no reason for others to know who placed those contracts.
2. Crypto is a unique and new player in banking and securities. Specifically, the ledger is open to all and very transparent. In order to know which wallet owns what and how much, you need to be able to look at the ledger to confirm. This design is intended to prevent fraudulent transactions. Banks and brokerages don't keep an open ledger, they keep an extremely private and heavily secured database.
3. People are heavily disincentivized to disclose this information. See: Epstein, and the people and companies that did financial business with him. Openly telling the public you sold stock to someone who trafficked underage girls isn't good for your reputation.
4. Futures in this case are similar to stocks. It's literally a market: You put in a bid to buy or sell something, someone else accepts, and both parties agree to a contract. Digital trading systems match buyers to sellers, and does so on behalf of both parties. If you trade using Fidelity and place a bid for 1,000 units of Corn at $500/unit, you might end up buying from Chase. Even if the transaction is in-person, the trader may be acting on behalf of someone else. In the case of the $580 million deal, it was spread over 6,200 contracts.
5. The STOCK Act of 2012 was supposed to prevent this, at least for members of Congress. The PELOSI Act that's currently introduced for voting is supposed to further prevent this.
6. To my original point: the problem isn't that these trades occurred. The problem isn't even that it's almost certainly insider trading. The problem is that government secrets are being leaked. The authors argument is that it almost certainly won't be investigated and any attempt at investigation will be blocked, because the level of corruption in the current administration is such that the sale of state secrets for others to profit off of is permissible. In fact, they can brazenly do it in the open while still ensuring that their privacy will be kept because of #1.
Unfortunately it’s mostly wrong. And the reason it’s not public is much simpler. These markets are both standard contracts on CME matching engines.
The CME uses a system where orders (and fills) are entered via a direct TCP/IP connection between the trading system and the exchange. There is no opportunity for any system besides the trading systems (computers, switches, etc) on each side to see the order.
The CME then distributes the market data to a variety of paths in the form of price updates to an order book. That is the price and quantity of buy and sell orders at each price point. While these data feeds aren’t public, the cme gets paid for the data, it is widely disseminated, but it’s had the trading system identification removed. This whole story comes from people with market data contracts observing these feeds.
There are other exchanges that provide more information about who is entering orders but infrequent but large participants don’t like this because it allows the market makers an information advantage against them. And large block participants tend to correlate with real economic activity (oil producers and consumers for instance).
If a market maker knows an order is for one of these participants they can presume the size is going to be bigger and thus charge more for liquidity (in the form of of widening the price on their bid/ask spread offerings). Half the job of a modern HFT is predicting this sort of thing.
So the cme keeps it this way so one set of their market will keep using them. It’s part of the balancing act a two sided exchange has to navigate.
While I agree with the risks of DA/stalkers getting that data, this data is not known for being well protected against LoveInt. Quite the opposite it is usually sold on grey markets.
A mix of public (city councils) and private (think HOAs that then donated access/equipment to the city) contracted with Flock in the past few years.
The questions of exactly who, when, and why, are very muddy especially with the HOAs who operate rather privately.
> Investing in community college programs is one route since they are essentially a business.
A 501c3 nonprofit with pretty stringent requirements (accreditation, reporting, transparency), but yes a business nonetheless.
> The main problem with the CCs is that they are very corrupt, have been issuing students worthless degrees.
Could you elaborate on what you mean by corrupt, and issuing students worthless degrees? Graduates from the program I direct at a community college are generally earning 120-160k in IT around 2 years after graduation.
> The colleges' goal is reaching a "graduation quota," and not "employability."
Universities yes, community colleges I would not consider this an accurate statement. At my community college our CTE programs (job training) are explicitly evaluated on student salaries as well as how many are actually employed in the industry after graduation, usually within 18-36 months time. It's actually two of the few metrics that are considered "high value", as in 2x the other value of other metrics like graduation, enrollment, retention, revenue-per-student, etc.
Community College professor here, in the midst of leaving my community college for a full university.
Let me dissect this article with uncompromising scrutiny:
> "...have over a million openings in critical jobs, emergency services, trucking, factory workers, plumbers, electricians and tradesmen.”
Maybe because for 30 years America sold the idea that you need a bachelors degree to do the majority of these jobs, while simultaneously implying that you only needed 2 years of vocational school? A lot of these require extensive apprenticeships and experience (576 classroom hours, 8,000 experience hours, passing exams for a journeyman electrician license in Oregon). It's absolutely not "Go to school for 2 years and get paid $120k."
Furthermore, most of the trades are brutal on your body, mind, and lifestyle.
> "we don't have trade schools anymore"
We do, and we do our best to train students on only the absolute necessary skills that get them the job and working as quickly as possible. Corporations stopped meaningfully supporting them while simultaneously raising expectations. Major companies stopped most training and orientation programs or significantly scaled them back, passed the burden of training onto community colleges and trade schools, and now complain that our tools and techniques are out of date.
Ford does at my college this while keeping their name slapped on the auto mechanic's program because they helped start the program 20 years ago. Now they're upset because they're not getting the same returns while my fellow instructors struggle to teach on supplies that are 2 decades old.
> "What we don't have are enough young people with the literacy and math proficiency needed to learn skilled trades."
A lot of the K-12 complaint is the No Child Left Behind act and the effects of Common Core. Lots of throwing up of hands here saying "Well guess there's nothing we can do. We have all these high paying jobs that no one wants"
Wanna fix this? Eliminate No Child Left Behind. Actually invest in teachers, tutors, and the people making the impact. Stop calling teachers 'heroes', and give us the resources to actually instruct kids. Stop assuming a household with 2-3 kids, 2 parents that work full time (overtime in today's America), are barely making ends meet, and have no extended family to help kids with homework or tutor, are going to somehow do extremely well.
In fact, we have loads of papers that demonstrate that math scores and grades are pretty tightly correlated with parents'/family ability and availability to help kids with homework. Maybe have parents work less so they can tutor their kids more?
> "Workers who struggle to read grade-level text cannot read complicated technical manuals or diagnostic instructions."
They don't have trouble reading grade-level text. This is a complete misunderstanding of what those tests evaluate. More importantly: If they're struggling to read those complicated manuals or diagnostic instructions, maybe it's because most manufacturers eliminated a lot of the repairability of cars in the past few decades and scaled back their service manuals? Maybe invest in technical writing again?
> They were passed on with inflated grades
Because you stopped hiring anyone with less than a 3.0-4.0. If a teacher's job is to get a student a job in the trades, you won't hire them because their GPA is poor, and we get fired if too many students fail, guess what we (instructors) are going to do?
> "If they can’t handle middle-school math they can’t program high-tech machines or robotics, or operate the automated equipment found in modern factories and repair shops."
Also not correct, and a gross misinterpretation of what the national exams show. Most students can do most math with a calculator just fine, mental math not so much, but it's rare to be in a shop without some kind of computer or calculator nowadays. If you want people who have completed a 2 year trade program to be able to competently do calculus, robotics, PLCs, and program, you need to admit that the job requires far beyond 'middle-school math'.
> ""Servicing an electric vehicle requires interpreting data flows, troubleshooting electronics, and following precise, multistep instructions." It's not a job for "grease monkeys."
Here is the crux of the problem. All of these are needs that are way beyond a standard mechanical technician's toolkit. You need them to dual train as electrical engineers and mechanical engineers with notable expertise in 12/24v and rather high voltages for EVs. You don't want 'average technicians' for 120k, you want dual-degree mechanical and electrical engineers to work for you for less than their going market rate. If your toolchain requires more than an understanding of ODB2 (or 1-2 device) readings and a solid understanding of vehicular operations and what commonly breaks, then you've spent too much time making your products unrepairable and obtuse.
A trucker (CDL) is the only one without extremely high apprenticeship requirements.
From there, who thought that you'd need a bachelors to be a barista and pour coffee? It's not just about the raw requirements, but about the competition you face in the job market.
Finally, more and more field service technician and electrician, HVAC, roles that are traditionally GED/2-years only, have extremely high experience requirements, and most are preferring people pursuing or with a bachelors in electrical engineering (electrician, HVAC), mechanical/fluid engineering (plumbing), or similar. Earlier this year I was in a remote-ish location (about 100 miles from a major city) and we had an electrical fault that legally required a licensed electrician to repair. We had multiple electrical engineers offer to help who clearly knew the problem and how to fix it according to code, but we couldn't let them touch it because they didn't have their J or better.
If you want to risk no degree and go for your 8,000 hours to get licensed (roughly 4 years experience for no/mediocre pay), go for it.
> who thought that you'd need a bachelors to be a barista and pour coffee?
"About one-fifth of underemployed recent college graduates—roughly 9 percent of all recent graduates—were working in a low-skilled service job" between 2009 and 2013 [1]. Two fifths of baristas have college degrees [2].
This used to be what people did before COVID. Now it's a hardcore seller's market in the US:
The car chip shortage caused the resale value of cars to skyrocket. In the past few years my 15 year old CRV, which just passed 100k miles, has gone UP in value according to my mechanic because the used car market is so bad.
Many car dealerships, already well-known for poor consumer practices, have straight up refused to sell cars to people who don't wish to finance because they can make a lot more money off someone who wants to finance. You used to be able to convince them you planned to finance, then buy the car in full at the last minute, and the salesperson would do so because of sunk investment. Not anymore, many of my friends have experienced this firsthand when shopping for a new vehicle.
Combine this with it being all-but-required for people in the US to have a personal vehicle and drive if they want a job ("reliable transportation" in job postings), new & used car dealers have the additional leverage of time pressure as people can and do get fired for their car breaking down.
What I see a lot of wealthy people do, on the other hand, is finance a luxury vehicle, drive it for 3-5 years, then trade in for a newer model. Used to be lots of luxury vehicles, like the Giulia Quadrifoglio, with 20-30k miles for 30-40% of the original vehicle cost (25k-35k) which is an incredible deal for a luxury brand. This isn't really a thing anymore as the cost of a new car skyrockets.
Depends on what you need. For general knowledge or broad exploration, there are plenty of general education courses you can take at community colleges. However for specific knowledge that requires lab space or equipment you can't get at home, you generally need to enroll in a degree program nowadays.
If you simply disable OneDrive without correctly uninstalling, the system will blast notifications at you with an ominous warning "You could lose data your system isn't backed up!"
PowerShell and most CLIs are terrifying to non-technical people. Literally Here Be Dragons. The layperson might be skeptical of a YouTuber telling them to run a dodgy script, but in the age of "delete system32" people sure as hell aren't going to run a command as admin that a user on a random forum recommends they run.
Stuff like this is why I have moved all of my systems except my gaming PC to Linux.
Edit: no seriously look at this notification https://learn-attachment.microsoft.com/api/attachments/f5907... Grandma absolutely does not understand what that means, She just knows she doesn't want to lose photos of her grandchildren.
reply