Yeap, you now get logs from all stages of the boot process, and you notice logs going bad or getting manipulated. It's a huge step forward for Linux, especially for people that look at logs every day. They can finally trust the logs to be correct and complete.

And it is not even a change: I have never had real unix servers with plain text logs in 30 years working with them. Proper computers have always stored logs in databases or whatever. That's actually a legal requirement in many parts of the world for many kinds of servers.

Just use a distribution that uses the init system you want to test by default. The init system is a pretty central piece of infrastructure in todays unix, ditsributions need to invest a lot of effort to make it work well. The default init system always works better than any other options a distribution offers (if they even over options).

There are significant security/reliability improvements all over the place. E..g. logind actually works as promissed, which none of the clones do as they can not have the necessary infrastructure that is provided by other parts of systemd. Or udev using systemd-pid1 to start services: That fixes a well documented problem with udev starting services itself -- which most non-systemd distributions do till today. Problems do notngo away by ignoring them.

Checkout non-systemd distros, most of them still use group based access to devices a user needs to run wayland. Of course that does not matter at all for X11, security is so poor there anyway.

These comparisons assume the systems do the same thing, resulting in a comparable system once everything is up and running.

That is just not the case. Systemd does a ton more wrt. security, hardening and reliability. E.g. it has udev integrated and not running "stand-alone" since it tends to lock itself up when it needs to start long-running services in response to HW showing up. Yes that happens rarely but there is nothing you can do about it. Logind locks down permissions to HW way more tightly than the "forks" that have the same problems that the pre-logind system had -- and that led to logind getting developed in the first place. Lots of sandboxing options are built into systemd and widely used to run services (I rarely saw any sandboxing elsewhere so far out of the box), measured boot is pretty much a systemd-only thing at this time, ... .

How does systemd dare to provide useful functionality? It should be just as useless as all the other inits that nobody ever felt the need to depend on, simply because they do nothing interesting.

You can just build a way more functional and secure system with all the cool features systemd provides.

The security theater is strong in this one.

"Nothing on this page asks for your faith.", immediently followed by a whole list of "trust me" (to manage the keys correctly, that the person notices the keys getting stolen, that she manage the entire signing process correctly, ...). No HW is used that could make it harder to steal keys, just some offline USB storage and a password manager. Every mainstream distro does way better than this, incl. Void linux this is based on.

No systemd with all the security improvements that brings either, but I guess that will bring in a small fan base:-)

I doubt they were the first to built such a system. The idea is pretty obvious after all.

They just are the first to give reporters a guided tour of the system and test accounts to play with.

You said Lennart is obnoxious and there is nothing like a big "that is not my problem" to not get your project adopted.

IMHO he does manage to get his projects widely adopted. His personality does not seem to have the effect you claim it has.

Systemd said on its first website that its a system management daemon. That is where the name came from. It was never supposed to be just an init system.

No need to buy out udevd, considering the guy doing that was on board with systemd from the start and made sure that systemd-init can make maximum use of udev and the other way around... you typically do want to start stuff in response to hardware appearing and disappearing. Now they can do that safely by just asking systemd-init to manage the services. They needed to run stuff thenselves before, which pretty often ended up blocking the udev daemon from recognizing new events... a quality of live improvement for everybody involved:-)

The rest has similar stories (only run network services when the network is up, start services only after the system clock has a sane value over starting the service and then adjusting the time at some later point (which some services handle really poorly), ... . There are some damn good reasons for the stuff systemd does. That bootloader is a pretty central piece in the image by the way.

Oh, and did I mention that anyone using or talking about non-systemd methods of doing things tends to get painted as "oh that's OLD and OUTDATED and OBSOLETE, just use the systemd way it's Modern™ and good"?

That is pretty much the only thing I can agree with:-) And that is because the systemd ways are ofentimes way more robust and able to deal with corner cases way better.

Considering how widely Lennarts code got adopted: You must be missing something.

Either you build a tool that works well with one OS or you build one that works poorly with several.

And it is not as if those other OSes are waiting for new cross-OS init tools: They either seem to be happy with what they got or want something tailored to their own features.

Ok, thanks for the summary.