And then the ship places all the passengers in stasis while it waits for its complement of small lemon-soaked paper napkins.

I can't say there are many times I've said "oh, if I could only run this on my computer...".

On the contrary, I say that to myself all the time. I hate using my browser as a mini-OS, with applications running in tabs inside the same window, with only one visible at a time. These days, everything seems to be available as a mobile app and a web version for desktop. And very often, the mobile version is nicer to use than the web version.

Sure, I could drag tabs out into their own windows, or install the site as an "app", but it sucks. I'd much rather have an actual app. The thing I'm most looking forward to in the next couple of years is Microsoft's MAUI framework, which will allow a single codebase to produce builds for Windows, macOS, iOS, Android and Linux. Then devs will have no excuse for not providing native apps across all platforms. Using the web to achieve this is terrible for usability.

I think you misunderstand the problem. If you build an iApp, you absolutely cannot accept payment for purely software-based purchases (e.g. unlocking a level in a game, or upgrading from the free to Pro version of an app) unless that payment is an App Store in-app purchase. You may have access to a 3rd party payment gateway which you'd prefer to use, but Apple will not allow you to make what it deems to be an in-app purchase via any outside means.

Take the argument that Hey.com's founders have just made: They don't sell subscriptions via their app, they don't sell upgrades via their app, they don't expose anything billing-related via their app at all - all of this is done via their website. But Apple has blocked the release of updates to their app, and is threatening to remove the existing version completely, unless Hey.com exposes payments via Apple's infrastructure (and only Apple's infrastructure).

Apple absolutely requires you to use their infrastructure, even if you have other alternatives available to you. They've made exceptions for physical goods, and for ride-shares, and maybe a few others, but if you want to monetize your software product, they demand their pound of flesh.

I've built an app which exposes in-app purchases as well as physical purchases, and I take direct card payments for the physical purchases at a 2.5% fee. I already have the infrastructure and the code support to handle purchases via an external payment provider. I actually had to do more work to support both in-app purchases and payment gateway purchases, but the app would not be approved if the in-app purchases didn't use the Apple-provided payment mechanism. I don't want to use Apple's infrastructure for payments at all, but I have no choice, so I have to pay 30% for some transactions and 2.5% for others.

If my payment gateway can make a profit from a 2.5% transaction fee, how can Apple justify 30% for performing the exact same service?

Yes, that's great. But when the instruction to ensure you're not uploading heic images goes out minutes before your exam starts over a tweet, you're probably going to miss it. Who the hell expects kids to be sitting on twitter minutes before they write an exam?

And for the great majority of those students who missed the tweet and submitted an heic, the site locked up until their time ran out, so they didn't have a chance to realise that they needed to convert their images to another format.

Maybe the special snowflakes at the college board need to learn to handle unexpected input gracefully instead of adding extra stress to kids who are attempting to submit an exam?

If you read TFA you'd see that the problem was that "The website got stuck on the loading screen until Bryner's time ran out."

Maybe if the site returned immediately with 'Unsupported image format, please re-submit as a .jpg or .png', you might have an argument. But when the site locked up and didn't allow you to do anything else then you don't know that you did something wrong so you can fix it vs the site is not responding and this is out of your control.

Two years ago I was tasked with building a mobile app that allowed uploading images from your camera roll. Worked fine on my old iPhone 5S which was uploading .jpg. Released to the client to test, and immediately got complaints that they were getting errors when trying to upload some images.
But guess what? Instead of just locking up and timing out or crashing the app, my backend realised that it didn't understand the image format, and returned an error immediately. And I could then look at the server logs to see what was causing the error, realised that they were uploading .heic files, and added in support for it.

Everyone here is blaming the students who are already under stress due to writing, and now being unable to submit, an exam. Very few are rightly blaming the developers of the system for:
a) not testing the very common use case of using an iphone to upload your pictures
b) not handling what should be an expected failure case of users uploading unsupported formats and instead of returning an error, locking up and timing out the test

Lots of those students would have been able to convert their images to .jpg or .png if they knew that there was a problem with the image format in the first place. And before the inevitable 'students should read the instructions before taking the test' nonsense responses that are already filling up this article, consider:

"One senior, who asked to remain anonymous to avoid repercussions from school, said that the College Boardâ(TM)s tweet went out just a few minutes before his Physics C test began"
and
"The College Boardâ(TM)s tweet went out just a few hours before Spencerâ(TM)s scheduled exam; he doesnâ(TM)t have a Twitter account and didnâ(TM)t see it."

Information like this should not be sent out over a damn tweet. It should be in big bold red text as part of the instructions before starting the test.

The college board is 100% to blame for this, and those of you who are blaming kids for their mistake, having years or decades of hindsight and experience in the IT field, should be ashamed of yourselves.

Would you mind sharing a link to your store? I'm in the market for a decent pair of earphones at the moment.

The same people who turn off the telemetry are the ones who complain about crashes or bugs that would have been picked up and fixed by said telemetry.

This whole telemetry nonsense started because someone noticed that MS outlined that certain keystrokes may be included in telemetry to improve typing predictions and autocorrect, and the M$ brigade turned that into "Microsoft is stealing your passwords and hacking your bank account and stealing your CEO's corporate secrets and giving your dog cancer".

So now, this whole telemetry story is the biggest case of FUD in the past few years, ironically promoted by those who claim to be against FUD, just because they have a pre-existing bias against the evil M$.

Nobody can explain why they hate telemetry so much, aside from "Windows 10 is a keylogger!!! Microsoft is stealing your passwords!!!" and "My privacy!!!". And then will happily load up dozens of web-based Google services in their Google browser and smugly transmit 10x the telemetry instead.

I would have no problem paying for a Disney+ subscription if it were available in my country. I'd probably even keep it indefinitely since I have an 11yo daughter.

The problem is that they've made the Mandalorian a Disney+ exclusive, and Disney+ is only available in a handful of countries. If they'd license it out to say Netflix in the countries that Disney+ isn't available, then it would be a net win for everyone - Netflix pays a huge amount of money to Disney to stream Mandalorian outside of Disney+ regions, and end users pay Disney a huge amount of money in regions that it is available.

So until D+ is available in my country, what are my options?

I'm vegetarian. Have been for almost 40 years, although I've tried eating meat in the past few years to make things easier at home... I just don't like the texture (and often the taste), but that's more due to being used to being a vegetarian.
I tried an Impossible burger a few weeks ago. Had the same impression I've had from eating actual meat - don't like the texture and not that fond of the taste either.
I suspect many vegetarians will feel the same way.

This isn't targeted at vegetarians/vegans - it's targeted at meat-eaters who want to eat less meat for whatever reasons.

Actually, MS found that disk IO in Linux running under WSL was slow, not that Windows' IO was slow. WSL1 doesn't have a Linux kernel at all, it is just a wrapper around the Linux kernel interface that redirects Linux syscalls to Windows ones. So it's literally impossible for WSL's IO to be faster than native Windows, because WSL is using the Windows kernel for IO behind the scenes. It's pretty obvious that the same binary running on WSL would be slower than running directly on Linux, because WSL is emulating the Linux kernel, which obviously has additional performance costs.

The gvfs project had absolutely nothing to do with WSL. It is, after all, a virtualised filesystem for accessing the Windows codebase, which would be pointless to checkout on a Linux machine, since it's Windows source code. Since it seems you don't know much about what you're talking about, here's a refresher: when you clone a git repository, you get the entire repo, including all of its history and changesets. A repo representing the entirety of the Windows operating system, spanning the past 20 years or so of commits, is huge. Simply cloning the entire thing would take a very long time and a lot of space. Hence they implemented a virtualised filesystem which only fetched the files that you needed as you needed them, instead of making them available from the start. Since it's unlikely most of the Windows team needs to look up revisions from a decade ago, having all of those changesets available locally isn't all that useful, so when you actually do need to look up something from a while back, it's fine to have to wait a little longer for those files to sync.
Additionally, as a result of their migration to git, they contributed back a number of performance improvements into the git codebase as well, since they began encountering performance issues with git that others would not have experienced, since not many organisations have a single repo of that size.

And before you claim that Microsoft is just too dumb to manage a git repository, Google built their own source code repository, and a similar virtualised filesystem for their own codebase.

I open new sites in tabs, because I read one site at a time, and that occupies my focus. However, if I'm working on a dev project, I will probably want to see documentation, source code, and have a console window open all at the same time. I would probably want my browser to be visible while my terminal is open so I can type out commands from the stackoverflow page I'm referencing. All of these apps are related to the project that I'm working on, and I want to see them all at the same time because I'm accessing information from all of them at the same time.

If all these apps were grouped together in a virtual desktop for 'dev project 1', then I wouldn't be switching between this VD and another one all the time, because I'm currently working on dev project 1 and will be focused on that for the next hour. I also don't want to be distracted by having a dozen other windows in my alt-tab list when they're not related to the project I'm currently working on. But if I need to switch to dev project 2 suddenly to attend to a support ticket, then I'd just have to change desktops and my support environment would be ready and waiting, with the ticket details open visible in one part of the screen while the code is visible in another, so I don't have to keep switching tabs between the two to keep looking up details of the issue I'm trying to investigate.

Even if sets aren't a full desktop, they're still one-app-visible-in-the-grouping-at-a-time. And I realise my perspective may be biased as I'm a developer, but my non-tech wife keeps multiple apps on-screen at the same time on her 13" laptop very often while doing admin work for her online store.

It makes good sense to group instance of the same app together in tabs - a file explorer or command shell for example. But not when your work involves working with multiple apps at the same time.

Tabs are really bad for productivity because you have to keep switching between them, instead of being able to position multiple windows around the screen and see them all at the same time.

What they should have done was leverage off Virtual Desktops to be able to save a Virtual Desktop configuration so the same applications open whenever you load up that Desktop. So I could have a VD configuration for Admin which automatically loads up mail, accounting package, etc, a second configuration for Dev Project 1 which opens up Visual Studio with the right projects loaded, SQL Server Management Studio to the right DB connection, documentation in Word or whatever, then a separate configuration for Dev Project 2 which opens its own set of apps.

The Virtual Desktop becomes the app workload grouping, and the taskbar becomes the tab control. The only thing they needed to add was the ability to save the VD state or configure a new one, and they'd have a much better solution that users wouldn't be confused by. And users would be able to interact with their apps in the same way they're used to already, instead of being forced into a tab-centric workflow.

I think you meant: until someone ports it to EMACS

No, it rather means that the browser code is generic and correct. If an element is overlaid over a video, then that means that there could be content in it that would need to be overlaid on the video during rendering, meaning the video content now needs to be software rendered and can't make use of hardware acceleration, which is the issue at hand.

The browser can't just pretend the div isn't there because it's empty, because in today's javascript-driven world, any element can be changed at any time, so you can't drop empty elements even if it looks like they're doing nothing.

Given that thus far the div appears to be useless, and that YouTube refuses to remove it, it appears that it's been added purely to sabotage benchmarks for other browsers, since Google can optimize it away, knowing that it's useless. But no other browser can do that, because, given Google's behaviour of late, as soon as Edge starts dropping this useless element, Google will start inserting something into it, and then claim that Edge isn't rendering their site correctly.