So, re-inventing inetd/xinetd. And process-based services are old hat and slow. Event-driven is increasingly important. Which requires resident event handlers.

There are numerous other applications of ssh and other services beyond your and Poettering's 'cloudy' view of the Internet. Please stop screwing with them and dismissing them with a hand wave. Like systemds screwing around with event handling processes.

I can vote with my donations.

There is no such word as gullible.

I don't think that would matter in practice. This law wouldn't let them specify what *news* is allowed, only what news sources, and there would be a huge stink if they tried to block the major real news outlets. They'd like to, I'm sure, but I really doubt that they'd succeed.

Russiagate was absolutely not "proven false". Mueller's report and both the House and Senate reports (from committees led by Republicans) thoroughly verified it.

... Boeing will whisper a deal in the PMs ear. Offering to sell them some garbage in return for shutting down the Canadian program, cutting up the existing reactors and throwing away the blueprints.

Produce ship-sized SMRs. Problem solved.

Maybe they were counting on the pushback. The Feds step up and "suggest" that they turn the encryption off. Or things will go badly for them. They do, but it leaks out. Now the public is pissed. And it's getting near election time.

Are you sure there's only one? (Yeah, because you defined the problem thusly.) It's a good logical thinking puzzle, but nothing to do with the real world problem at hand.

But you haven't substantiated "the majority" other than staring into a bin and guessing about actual numbers. There's one red ball based upon your example because you put one there. You didn't define the ecosystem of real world use cases. And Poettering is even worse. Given his misunderstanding of the memory and library management of Linux utilities, I'm not believing his estimates. Even worse, his example of sshd is bizarre. Is he really suggesting that we not leave sshd running on systems that we infrequently contact? What? Am I supposed to drive across town and start it on a remote system when it's needed?

Well, not just one young arrogant engineer, also most of the maintainers of the major Linux distros in the world.

If it's really a bad idea, the blame doesn't really fall on Poettering. Many young, arrogant engineers have built things that were stupid, and their things got ignored by the world. Some smaller number of young, arrogant engineers have built things that were stupid but were able to convince their PHBs that they weren't stupid and they got deployed. I don't think that's how I'd characterize the leadership at Red Hat (I never worked there, but I have good friends who did), but let's suppose that they were clueless and that's why they deployed Poettering's stupid idea.

But then how do you explain why so many others looked at it, experimented with it for a few years, and then decided to adopt it, and even extend it?

The systemd opponents are loud and forceful on social media. The people who actually build the systems, however, disagree. And It's not just one or two groups who are somehow beholden to Poettering, nor is it people who don't know anything or have no technical stake in the decision.

You might want to consider whether you're living up to your nick here.

I don't personally care that much. I find it mildly annoying that the old scripts my finger muscle memory still wants to type by default don't always work... but honestly I rarely need them any more, because my systems Just Work. And I have to consider the possibility that systemd is part of the reason Linux requires so much less maintenance than it used to. There are multiple contributors here. A lot of it is that drivers have gotten a lot better and other aspects of the system have matured (like the audio subsystem :^)).

But given its broad adoption by nearly all open source and commercial Linux distros, Occam's razor says that it's probably better than sysvinit. Or BSD init. Or Upstart. Or OpenRC, or... <insert favorite system manager here>.

How can you tell how many red balls there are in the bin if you don't properly sample its contents?

Worse yet, if you are one of the blue balls, you can only see the adjacent 12 balls. If they are all blue, you are missing a significant, albeit a minority, of the bins contents. If we dropped 8% of a system's capabilities each revision cycle, pretty soon there wouldn't be much left.

Which still results in the bus being delayed as the cops wrestle with a bus rider having a mental crisis.

We're doing that here. What we need is plexiglass enclosures for the mentally stable passengers as well.

FTFY.

From TFA:

I stopped right there. How about running ssh multiple times per day? Sure, Poettering may never forward X11 over ssh from his laptop. But his complete lack of understanding of various *NIX ecosystems just demonstrates how unprepared he is to make important system architectural decisions.

Why? How often does a non-pro user ever have to dig into init system settings?