![](https://upvote.au/pictrs/image/84065ed2-43b1-4ab2-8678-dfcd7e760a7b.jpeg)
![](https://lemmy.ml/pictrs/image/q98XK4sKtw.png)
Do they use the BSD userland instead? Interesting…
I think Alpine uses Busybox, but it’s feasible for a Linux distro to use BSD coreutils. Not sure if any do that, though.
Aussie living in the San Francisco Bay Area.
Coding since 1998.
.NET Foundation member. C# fan
https://d.sb/
Mastodon: @dan@d.sb
Do they use the BSD userland instead? Interesting…
I think Alpine uses Busybox, but it’s feasible for a Linux distro to use BSD coreutils. Not sure if any do that, though.
User agents were commonly used for the wrong reasons - fingerprinting, sites that block particular browsers rather than using proper feature detection, etc. so I’m glad to see them slowly going away.
Interesting… I didn’t realise Skylake isn’t supported. I agree with your comment. I thought people were talking about much older equipment.
TPM 2 has been around since 2015ish and I wouldn’t be surprised if Windows starts relying on it more heavily. A lot of businesses have already required employees to use computers with TPM 2.0 for a long time, and enterprise use is a big focus for Microsoft.
not counting systems that use the Linux kernel but aren’t considered a traditional GNU+Linux desktop.
Does that mean you don’t count Alpine towards Linux market share? It mostly doesn’t use any GNU stuff.
You can also compile the kernel with LLVM instead of gcc, use musl instead of glibc, and use BSD coreutils instead of GNU coreutils.
User agent strings are frozen these days, at least in Chrome. They still have the browser major version and OS name at least, but Windows will always report Windows 10, Android will always report Android 10, MacOS will always report 10.15.7, and Linux is just “Linux x86_64”: https://www.chromium.org/updates/ua-reduction/
User agent strings are essentially deprecated and nobody should be using them any more. They’ve been replaced by User-Agent Client Hints, where the site can request the data it needs, and some high-entropy things (ie fields that vary a lot between users) can prompt the user for permission to share them first.
The thing is that most Windows users don’t care and will continue to use it. People like you and I know about the benefits of Linux, but sometimes we overestimate how much regular users care about the OS they’re using.
Forced restart for software updates
If anything, they’re moving in the opposite direction. Windows Server 2025 is going to support hotpatching, which means that system updates can be applied without needing to reboot. Not sure if the technology will come to consumer Windows though.
Require new CPUs and motherboards / hardware, ignoring the market for old computers.
How long do you expect legacy hardware to be supported for?
EV certs used to make the address bar green, so there was a tangible benefit to them. These days, they look exactly the same. It proves that the ownership has been verified, but if no users even see that, does it even matter?
If you run a website: Paid SSL/TLS certificates. Free ones like Let’s Encrypt and ZeroSSL are just as good, and can be automatically renewed.
I hate that these commercial providers are the first thing people think of when they hear “VPN” these days, rather than the actual main use case for a VPN (connecting to a remote network, like a work network, from another location).
AirVPN are probably the best. They’re independent, more transparent than the other providers, and support port forwarding.
Probably ozempic, since people going off it immediately balloon back up
Doesn’t that mean that it’s actually working, though? It has an effect when you’re using it, and the effect goes away when you stop using it.
Also the version for weight loss is called Wegovy.
In E2E tests you should ideally be finding elements using labels or ARIA roles. The point of an E2E test is to use the app in the same way a user would, and users don’t look for elements by class name or ID, and definitely not by data-testid.
The more your test deviates from how real users use the system, the more likely it is that the test will break even though the actual user experience is fine, or vice versa.
This is encouraged by Testing Library and related libraries like React Testing Library. Those are for unit and integration tests though, not E2E tests. I’m not as familiar with the popular E2E testing frameworks these days (we use an internally developed one at work).
In an alternate reality, we’d all be using JSSS, which was even worse. https://en.wikipedia.org/wiki/JavaScript_Style_Sheets
Why will it be better in just a couple months?
Explicit sync. It’ll fix most of the issues with Wayland on Nvidia CPUs. Wayland landed support for it in April, and Nvidia recently released a beta driver that supports it. I think every graphics driver will implement explicit sync eventually, since it’s a lot better than implicit sync.
Some great information about why it’s important here: https://zamundaaa.github.io/wayland/2024/04/05/explicit-sync.html
It’s an immutable/atomic version of Fedora: https://fedoraproject.org/atomic-desktops/silverblue/
My understanding is that the core system is immutable (read-only) and major upgrades essentially just swap out that whole layer. Updates are atomic, meaning the entire thing either succeeds or fails and you can never end up with a broken half-updated system. UI apps all run using Flatpak.
I’ve never tried it though!
This is the case with a lot of apps that follow SemVer, even though it’s not an official part of the spec. It’s not specific to Rust.
The other common thing I see is that if it’s been at 0.x for a long time, the minor version number eventually gets “promoted” to a major version number once the app is stable. For example, React went from 0.14.x to 15.0.0.
The “automatically includes a hashtag with new posts” link goes to the wrong PR.
What is a “top” story on Lemmy, given everyone subscribes to different communities? Is it the most popular across all communities?
There’s a good explanation about that here: https://webaim.org/blog/user-agent-string-history/
The issue is that a lot of sites used the user-agent to determine if the browser supported particular features (e.g. show a fancy version of a site if the user is using Netscape, otherwise show a basic version for Mosaic, lynx, etc). New browsers had to pretend to be the old good browsers to get the good versions of sites
This is why getting rid of the user agent is a good thing. Sniffing the UA is a mess.