It wasn't going to go any other way.

I'd delete the hysterical if I hadn't already marked it "solved". It appears there are a few for whom no amount of time will ever allow emotion to subside and reason prevail.

In 1973 I was working in a music store and we had our typical blowout Washington's Birthday sale with slashed prices (and a few bait and switch shills) and I got a phone call from a lady inquiring about the $100 USD drum set which she was very excited about since her husband "has always wanted a drum kit but couldn't afford one". She further told me she had been searching around for years and it seemed $100 was far and away the lowest price she had ever seen... was it a good set? I told her the truth... "It's more than adequate for a student".

Then she asked "Where is it made?" I replied "Japan, of course" and her response was "Oh darn... that's too bad then. I could never buy that drum kit for him if it was made there". When I asked why that was a problem she said "His Dad fought Japan in WWII and he won't buy anything made there". I told her that it was highly unlikely that any of the people that made those drums were even born yet by 1945 so who is he against?... a group of people with certain ideals, agendas and concepts or every person ever born on a particular island into infinity? Maybe I should've been more clever and point out he could imagine Japanese faces while he beat on the drum heads, but I'm just not a sociopath and won't play into such ugly delusions.

It appears some people never see the world in light of Peace. For them it is always a State of War.... endless war with only one possible absurd outcome - The Last Man Standing.

1 members found this post helpful.

True BUT taken individually on their own cost/benefit analysis not just because they remind you or anyone else of Windows... or maybe we should abandon Zeros and Ones?

This thread is marked "Solved" and while as OP I don't mind if people continue to post in it, I do expect them to actually contribute something of value, whether for or against. You obviously didn't watch the video, have nothing new to say let alone any specifics to contribute so unless you do, please refrain from mere "beating on hateful drums".

I'm a little sceptical about the shorter boot times. It seems that systemd is only faster because it defers much of startup until later. When Slackware finishes booting I know everything is ready. With systemd, the startup of half the stuff may still be waiting on socket activation.

Windows 10 uses the same trick. The login screen shows up very quickly, but if you type your password in as fast as you can and then quickly try and launch an application from the desktop you sit there waiting while the rest of the boot is still completing in the background.

In short, you're being fooled into thinking the boot has finished, when it hasn't.

1 members found this post helpful.

Shorter boot times while not an original design goal are commonly touted as "frosting on the cake", but the nature of parallel initialization still depends on everything that needs to be loaded actually loading before boot is complete. While, as the linked video speaker points out, systemd has some potent tools to monitor what has loaded and what has yet to be loaded, one single failure can throw a huge monkey wrench into the process, or at least it did 4 or 5 years ago in my experience. It was not at all uncommon for both Ubuntu and OpeSuSe to hang for long times "circling the wagons" to defend against sequence issues. This may have depended to a degree on older, slower hardware but the same hardware was in fact faster to boot in serial mode in Slackware 12.2 and 13 I was running during the same period.

So the truth seems to be that systemd CAN, has the potential, to boot faster but that isn't always the case and who really cares? How often do you reboot with Linux? If you do even boot a time or two each day how long is that? 5 minutes tops on a really slow machine? These days Slackware boots for me to runlevel 3 login in less than 30 seconds (around 27 to be exact). Big deal? N-n-n-n-n-n-n-not really.

And I helped!

2 members found this post helpful.

Your willingness to paint your opponent as operating purely on emotion and spite is intellectually dishonest and basically trolling.

You don't acknowledge / agree with the reasons they do, that doesn't create the need for straw man and ad hom. Tis the thread / topic sure, but there are companies that are actively engaged in trying to destroy our software ecosystem.

It isn't fiction, it isn't hatred-- this is about the actual survival of something that took decades to create. It's increasingly well documented, as are the zero-sum tactics of the people being mistrusted. But whatever.

These are monopolies. You're not talking about forgiveness years later, you're talking about surrender in the middle of an ongoing campaign against our own development. Their actual model is to systematically destroy the competition, which they're still doing right now. Ask Nokia. Ask Eric Lundgren.

Not wanting software lock-in and being against it is a perfectly reasonable stance for people that spent years getting away from it. It's a very practical matter, which you insist can't be a practical matter just because the people involved also don't like jerks.

They're still doing the same things they did 10 years ago. To respond to your silly metaphor, then by all means, forgive people from countries involved in past wars. But maybe don't hire the same individuals who killed the guys fighting alongside you to babysit your kids, especially if those individuals are still involved in the occasional (publicly documented) assassination or kidnapping.

A more accurate metaphor would have nothing to with a long-forgotten war, and more to do with escaping an organised crime family. How exactly would "forgive and forget" factor into that scenario?

Sure, that's what you did.

3 members found this post helpful.

To be honest, I have never had the feeling that systemd boots noticeably faster. I have never timed it, but as far as I can tell, neither Debian, nor Ubuntu all of a sudden booted any faster after they migrated to systemd, and I don't really care anyway.

My only real contact with systemd was when my desktop computer didn't want to power down when I attempted to shut down Ubuntu or Debian when running specific kernel versions. The issue had something to with USB hardware that didn't get properly disconnected. The computer seemed to try and power down alright, but then suddenly woke up again and rebooted. The workaround consisted of a small script that had to be run just prior to the power down, and it was surprisingly straightforward and elegant to implement under systemd.

Having said that, I'm somewhat suspicious about the way the systemd developers seem to handle security in particular. I'm not convinced that I should trust them with my computer. It looks like systemd mostly works fine, but that's what I thought about Windows, too, as long as I didn't know any better.

As far as I'm concerned, the jury's still out on the systemd question.

Eric Volpe, an Ubuntu user and professional sysadmin (and incidentally the guy behind Teletweety) has measured systemd boot times, and he said it took a lot longer to boot his Ubuntu 15.04 system (with systemd) than his Ubuntu 13.10 system (without systemd).

Unfortunately that isn't quite an apples-to-apples comparison. 15.04 might also be trying to do a lot more useful work during bootup than 13.10, so there is still room for doubt.

When I used Debian for a few months before coming to Slackware, the boot time was significantly faster on my laptop, for what it's worth (Thinkpad laptop with HDD). It was kind of fun to watch it boot up and marvel at how fast it was, but ultimately not very important. These days I rarely shut down in the first place.

I'd respond to that, but then this thread would very rapidly become more vile than any of our past systemd threads. Have a nice day.

This was my experience as well, on Debian systemd was significantly faster to boot. This was my Debian boot time from on old post on the Debian forums:

I just did a trial restart on the same machine with the same hardware which now runs Slackware. The huge kernel took 34 seconds to boot to login, and the generic kernel [which is what I normally use] took 44 seconds.

So in one case the kernel was more than three times slower than systemd to boot, and in my normal use-case, more than four times slower.

Nevertheless, I still consider those times to be fast and in no way part of a consideration for moving from a SysV distro.

Three reasons why I don't use systemd ...

Code quality: goto, return, break and continue in PID 1 - are you serious? Meaningless variable names - are you serious? ...

Developer attitude: ignoring bugs in a security relevant part of the system - are you serious?

Violation of the KISS / Unix principle.

4 members found this post helpful.

It still probably couldn't compete with the vile spew that came from the systemd proponents when they took over Debian.

I mean, people don't want their systems tampered with by dishonest and immoral developers, and for that there is hatred-- not of the developers, but of the people who didn't want Debian taken over. Sure, that's fair...

There's nothing that systemd does that can't be explained by clinical narcissism and incompetence. There's nothing done to promote systemd that can't be explained by narcissism and institutional corruption.

Other than that it would be a mystery why people are so eager and cheerful to smear critics and defend sabotage. You know why people are so divisive over systemd? It's the most divisive project to ever get thrown into the middle of the free software ecosystem. (You might make an exception for the Linux kernel, but at least that's highly useful to both sides.) It's not an init-- it's a weapon, and it works as designed.

3 members found this post helpful.

There's nothing wrong with using goto, or break/continue in the right situation. I'd rather see a break statement in a while loop than see something completely incomprehensible spread over a bunch of convoluted if statements 10 levels deep.

Not that I'm defending systemd code quality, I'm not making any judgement one way or the other on that, just stating that the 'structured programming' dogma of not using those 3 statements, at any cost, is IMO quite simply just that: Dogma!

1 members found this post helpful.