From a health-of-the-Linux-ecosystem point of view, Gentoo developers seem more prone than those working for other distros to actually pay attention to things like broken build systems or test suites, and file bugs against them. This is especially the case for exotic arches that the developers of some software claim to support but don’t themselves build for (but it affects amd64 too).

I don’t think there’s anything preventing someone from forking it into an out-of-tree patchset or alternative module, if they’re enthusiastic about the feature. But I’m pretty sure that searching for someone to rewrite a non-core function that most people don’t care about is way outside what Linus and the various core maintainers want to do with their time. Normally open source is about people scratching their own itches.

But why they need to remove it?

Most likely maintainers are looking to reduce the maintenance burden produced by a feature that had very low uptake (assuming, as others are saying, that it’s only data journalling and not metadata journalling that’s being removed).

And also, pretty important, zero drama.

Not entirely true. The dev mailing list has periodic drama spasms. It just doesn’t tend to spread beyond that.

All they really have to do is specify that all new machines sold domestically have to be loaded with their tailored Linux distro and no other OS, and they’d eventually get rid of the majority of Windows machines. Enforcement of that should at least be possible, since they’d only have to check sellers and distributors of new equipment. Then just lean on domestic developers and/or software companies so that they don’t provide new versions of must-have applications (whatever those may be) for Windows to encourage the holdouts to move on.

It would never fly in a country without an authoritarian government, but China could likely get away with it. It just wouldn’t be instantaneous.

Sorry, but . . . Gentoo user. Meaning that one of the things I do is retain packages for software that suits me as long as I can still make it and its dependencies compile, because that’s one of the things my distro allows me to do. I’m sorry if your expertise ends with taking what’s explicitly offered to you.

Speaking as someone who doesn’t use any of these things, at least half the problem is the people who push these pieces of software. Hint: telling someone “oh, your preferred software is crap, use this instead” when what they have is already sufficient for their purposes makes you look like a liar and turns people against your message because they don’t like being insulted. There are examples in the comments here already.

There is a type of person who instantly jumps on new software because it is new . . . and y’know, that’s fine. We need those people, or software will never improve. However, most people don’t think that way and need to be offered a compelling reason to change. “This new software has this great new feature” may be a compelling reason . . . for someone who needs or can use that feature. “The codebase of the software you’re using is old and crufty and not attracting many developers” is not a compelling reason as long as there’s at least one person around providing needed patches. Even “the software you’re currently using is completely unmaintained” may not be a compelling reason if the software continues to work and doesn’t have any known major security issues.

I’m not sure what the mechanism is, but I discovered the issue by empirical testing a few months ago, and I’m still trying to sort it out and make the LAN visible to that VM. Granted, I haven’t been able to put much time into it, but I keep ending up down among the weeds of iptables/nftables, with all the documentation either out of date or not intended for a host using netifrc for network management (or both). I’ll probably have to ask on the Gentoo boards eventually.

Depends on your VM. qemu, by default, isolates each VM on a subnet. It’s a PITA when you actually need the VM to be able to see the LAN.

Barney. Is its mascot a purple dinosaur that everyone over the age of five hates with a passion? Because that’s the association the name calls up for me, and it makes me dislike the project just on general principles.

But what would prevent somebody plug in a USB with and just wipe my drive?

If they have enough access to plug in a USB key, they have enough to smash the drive (or the entire machine) with a sledgehammer. Or move it to another machine that they control and wipe and reimage it there.

(Shades of the xkcd with the crypto-nerd and the pipewrench: there’s always a non-technical, or less-technical, solution when it comes to security.)

You can install any desktop environment on any distro. For some of the more obscure DEs on some distros, a bit of hoop-jumping may be required, but it’s still possible. And some distros ship without any DE at all.

Technically, you can have a functional desktop without a DE at all, just a bare window manager (in X) or compositor (wayland), but that’s getting into advanced topics.

It’s a normal browser setting left over from the early days of the Internet, when most pages did little or no theming (or did it badly). My primary browser is Pale Moon, which exposes the setting as a dropdown under Preferences > Content > General > Colors. Firefox also exposed the setting on its settings page the last time I checked, although it may be a checkbox there. It’s also a boolean in about:config, browser.display.use_document_colors.

Chrome-family browsers may not have such a setting, unfortunately. The only one I keep handy is Vivaldi, which doesn’t have it but does have a “Force a dark theme on all websites” checkbox (under Settings > Appearance > Website Appearance).

I chose a slightly different way of achieving what that extension does, and long ago set my browser not to allow websites to choose their own colours. That does partially break a few sites that don’t handle their images properly, but overall it’s workable and gives me white on a slightly different dark colour for each browser profile.

The colour scheme. White screens are one of the things that trigger my migraines.

This in turn leads to my modifying several other visual aspects of the desktop, because a lot of widget set themes come with scrollbars that don’t work well with really dark colours. I also turn animations and automagic snapping and desktop navigation facilities off, because I find them annoying. Kindly behave like it’s 1999 and don’t do anything I don’t explicitly tell you to do, Mr. Desktop.

And then I embed a conky instance in the second desktop for general monitoring—CPU activity (composite and by thread), memory in use, disk space, network usage, temperatures, fan levels, and greediest processes all conveniently in a single view that doesn’t require me figuring out what I did with the window.

KDE provides a very large set of services. This makes it something of an elephant in the world of Linux desktop environments. Lighter DEs should fit in less than 1GB (that’s anything other than KDE and Gnome). I think the last time I checked my system with just the DE up and nothing open, it was using on the order of 200-300MB. However, the lighter DEs will also do less by default.

It should be safe to kill baloo, but as others have already said, you will see some tradeoffs in how file searching works if you stick with KDE. If I’m not mistaken, it may restart periodically to index new files unless you go to some effort to exorcise it.

Perhaps if you’re younger than the horse, it owns you instead.

That may have been the point.

I’d be more impressed if they stopped bundling random crap. It’s their unwillingness to use upstream’s dependencies and toolchains that makes Chromium a nightmare to package on Gentoo, and flatpaks are not going to fix that. They bundle clang, rust, and so many additional libraries that the “must keep bundled version to prevent breakage” list is almost 300 items long. This is not the correct way to ensure reproducibility, guys.

I’d be interested in tests across hardware of different ages (but as others have said, same/similar DEs/WMs/compositors as much as possible, please). While I have no proof, I’d guess that Wayland is making assumptions that would make it less efficient on some older kit.