This comment is ranty because I just stubbed my toe on this.

The fucking software repositories need to be current, goddamnit. Either provide a curated list of up-to-date software or don’t. Don’t pretend you have software in store just for the user to find out that it’s (literally, for fuck’s sake!) 4 years behind the current release.

Devices are configurable via software. If windows managed to “flip a switch” on the WiFi chip, it would affect Linux as well if it didn’t reset it on boot.

Oh, was it already time to reinvent the wheel? Again?

You could host your code, e.g. on Github, use Github’s code editor and only pull the code on your server.

https://en.wikipedia.org/wiki/Sarcasm, or maybe https://en.wikipedia.org/wiki/Hyperbole

Could it be that Steam overcounts the users? I mean if you have a Steam Deck, do you now count as a Linux user, thus diluting the Windows share, even though you’re still (also) using Windows?

Alpha from The Walking Dead.

Good point.

The logic is fine. If you rename the variable to isAdmin, it makes perfect sense. Either they are an admin, or they are not an admin, or the state is unknown (here expressed as null). If you want to throw another JS-ism at this, undefined could be assigned before the check has been made.

I regularly use variables like this. If users is undefined, I haven’t fetched them yet. If they’re a list, then fetching is complete. If they’re null, then there was an error while fetching.

Angular provides a utility to update its packages. If you naively update every single one manually, you’ll get errors because the library will be incompatible with itself. Maybe other flavors of SPA have update scripts, too?

ng update @angular/core@<the next major version> @angular/cli@<the next major version> Don’t skip versions!

For each problem you might want a package for, there are at least 3 packages, sometimes lots more. You need to sift through them in order to get the one that works and is maintained.

What I meant is that the “registering with the OS” part is sufficient. If the uninstallers suck, no operating system can do anything about that.

“OS level application management” is just the good old “Programs and Features” dialog where it simply registers the path to the uninstaller.

That’s what I took that to mean, too. And it’s sufficient. It allows the OS to provide a single point from which to uninstall all programs.

On windows, they have a registration scheme where installers log to a common OS level application management on what to run to uninstall.

Yup sounds absolute reasonable… Wtf?

What’s wrong with that? When I was cleaning out Windows recently, I was happy that I didn’t have to hunt down uninstall scripts in every program directory I wanted to remove.

First, and foremost, there shouldn’t be a weird thing jibbering up machines.

Consistency in the settings, especially in localization. For instance, in my Linux Mint start menu, I have “Settings”, “System administration”, and “System settings” (subtitled “Control center”). Now where do I look for a setting? Additionally, some or all setting from the “system settings” are available as standalone apps. Why?

In a similar vein, a run-of-the-mill distro is made up of lots of components, and it is not at all clear which is in charge of what. If I want to change hotkeys, who’s responsible for that? What do I need to google for? Drivers? Desktop environment? Some OS-specific settings app?

In general, there is always two or more of everything. Sound? pulse or pipewire. Which is installed? Which should be installed? Search the web and find literally every answer. UI widgets? Gnome, Cinnamon, KDE. Graphics? Nvidia, Noveau, PRIME, Optimus. The question “How do I make this work” always is a “well, it depends… actually… you’ll need to try, and if it doesn work, try something else.”

And which one would be better in the Debian/Ubuntu sphere, in your opinion?

Very efficient. I’m starting to feel this is what users see when they interact with websites, anyway. Only the cross should be “donation_dialog_button_close”.

If that’s the only error mechanism, sure. Exceptions in most languages tend to be relatively expensive, though, and most have a cheaper idiomatic way of returning error codes; you’d want to use those if they’re available, right?

I think not being able to get the current time from the system is very exceptional. And I think exceptional circumstances should act that way and not “look like” normal executions. To me, that means letting hell break loose, and not “silently” return a 1 instead of a 0.

By similar reasoning, “Exceptions in most languages tend to be relatively expensive” is a very weak argument. We don’t expect this error-throwing code to execute a lot.

there was a significant decrease in file fragmentation

is that relevant for SSD users?

With Linux servers continuing to have more CPU cores and more containers being loaded on each server, Huawei began noticing scalability issues within the EXT4 file-system driver code.

Simply put, the fallocate operations per container per second are able to come up significantly

So these improvements are for high-load scenarios, right? Casual users won’t notice any improvements? (Not to bash this contribution, just asking)