Amen to what exactly?
Amen to some shell script hack that try to reimplement everything that journalctl is doing ? ( Because no, i don't believe it's possible to reimplement all the function of journalctl with some simple one liner as
$ journalctl [options]

I believe too that journalctl would be faster then the same implementation with syslog that spawn each time grep, awk, find and whatever you want.

Everytime you call these utilities, a new process is spawned, the needed libraries by the program are searched, some stuff like i18n are setup, etc...

The same is true of journalctl, or any native executable. At this point you are only going to convince the ignorant.

If you call a lot of grep, find, etc... this is different since this process is repeated for each binary unlike journalctl which is one binary that allow you to do everything log related.

Okay, let me set a few things straight because there is enough crap floating around about what I've stated...

I have done enough homework having dealt with the travesty that happened with ArchLinux with systemd and the continued issues they still have with it's implementation. It's a very slipshod toolkit and getting stuff working at times was far much more a hassle than a blessing. With the amount of work it too to regenerate the systemd configuration files to work as I needed them, it took less effort to redraft sysvinit text scripts for my needs using a basic shell text editor like Nano. It's hard to argue against plain text in terms of simplicity.

Hacking/Re-scripting a shell script in plain-text isn't hard unless you're a complete idiot who can't read basic English documentation.

As far as IF I could work on sysvinit... I really wish I could contribute to the project, but I'm not a programmer. Just because a GNU project is open source doesn't give you or anyone the right to say, "if you want it then you code it". To me that's just disrespectful and condescending to the people who aren't programmers. I've seen a lot of projects adopt this mentality only to fail because the programmers who worked on the project started focusing less on bugs and more on candy features.

And Elvis again, you don't get the meaning behind my signature...

It means just because someone gives you a new shiny toy to play with doesn't mean it's always the best thing since the skateboard was invested. Often software that is presented as faster, easier, and better, ends up only being more bloated, less admin friendly, and overly complex to where people don't want to use it, but are then presented with the problem of, if you want to remove it, everything that now uses it has to be completely rebuilt from scratch.

That's basic system hi-jacking 101.

And yes, the3dfxdude said it best, Lennart took a pot-shot at BSD which he had no right out of arrogance and he's been rightfully backlashed for it. By taking a shot at BSD, he also aimed at Solaris, other UNIX branded OSes, and various other alternative UNIX-like operating systems saying that these OSes are pointless and meaningless against Linux.

And lastly, Elvis... for that immature and convoluted comment you made on my mother and her well-being... good luck with trying to apologize... You went over the line with that one, and for that, well... now you'll have to live with your mistake.

For someone who claims to understand the concept of UNIX and UNIX-like systems, you know very little if anything at all. Comparing Windows to UNIX is about as easy as comparing a Apple to a Cow and saying both are plants.

Linux does NOT gain any significant benefits of trying to be like Windows. Linux and every other UNIX and UNIX-like systems were NOT designed to be anything like Windows on any scale or level. If anything Windows should be trying to be more like UNIX and UNIX-like systems with modularity designs, minimalism within a system to where you can run one thing without needing to run everything else related to it, customizability within the installation process instead of the all-or-none approach, and the aspect that everything has to be ran by that dreaded svchost.exe and not through separated system services to use less resources.

Ever wonder why Windows Vista, 7, 8, 2008, 2008 R2, and 2012 require such a beefy system to work from and Linux systems like Slackware can run on i486 32-bit hardware from nearly 20 years ago, it's because everything doesn't have to be on the same service application bloating the system processing and memory requirements to Timbuktu and back.

You are the idiot here if you are seriously saying that a configuration file that is in the format:

key = value

is more complicated then a shell script...

I'm not saying that shell script is complicated, all i'm saying is that systemd configuration file is a lot simpler since it's not a programming language at all like sh/bash/whatever.

I was not gonna apologize at all, you totally deserved it.
If you don't want to be insulted, don't insult the work of other developper that I use everyday without any kind of proof besides subjective feeling.

I've watched the entire presentation sans the Q&A session at the end, and I made notes along the way. I found a total of one good argument in favour of systemd, and one half-decent feature that could just as easily have been implemented with SysV init.

I also heard about a great many features of systemd that actually exists outside systemd. A few quite nonsensical arguments were presented as well. I could post the summary and my comments, but I'm afraid it would be quite a lengthy post.

But most of all I heard about a project that has suffered scope creep to such an extent that it has spiraled totally out of control. The original goals have long since been abandoned, something Mr. Poettering readily acknowledged, and at this point the project has no clearly stated goals and no limits, which means it can never be evaluated as either a success or a failure. Or be considered completed, for that matter.

Mr. Poettering made a point of the fact that nobody knows how many other programs and how much functionality they will end up absorbing into the systemd project, and he seems to think this is a good thing. I disagree.

2 members found this post helpful.

Your basic shell script contains commands as you would type them on the command line. Nothing more, nothing less. The "commands" are usually regular executables, which run every bit as fast as any other process.

One of the differences between using a shell script and typing stuff manually is that scripts often contain variables that decide what the script will do and/or how it will do it. The variables are in the

key = value

format.

Why not make a Blog entry?

http://www.linuxquestions.org/questions/blog_usercp.php

They want systemd to be the building block of an OS.
Is it a good thing?
I personally think it is a good idea since everything in Linux seems cheaply glued together.
Even Greg K-H started to make a dbus implementation inside the kernel so everything will be more tightly integrated.
=> http://www.kroah.com/log/linux/af_bus.html

IMO, if you don't like the way Linux is turning, why not switch to a BSD flavor?
It seems to fit a lot more your "mentality" (no offense), then the Linux world.

IMO, if you don't even have a Slackware system running why are you so insistent posting to this thread in a Slackware forum? Why not spend your efforts in the Red Hat or Gentoo forums? It seems to fit a lot more your "mentality" (no offense), than the Slackware world.

Why do you feel the need to copy what I say and try to mock me with it ?
Did I hurt your feelings ?

We already have the OS. The question is: Should we throw out a lot of the existing components in favour of a new set of components called "systemd"?

Would you not agree that in order to to answer that question, it makes sense to examine what we gain and what we lose by doing so?
I would expect no less. He used to work on the project that has now become systemd.

Your "cheaply glued together" comment is just plain wrong, though. Have you examined the init system? Its simplicity is a thing of beauty. You can simplify it down to a single BusyBox executable if you like (many initrds do this), or create a startup system just as advanced as you like or need without ever losing track of how it works. And you can do all this with no other skills than the ones you need anyway to maintain the services running on that system.

Unless you run systemd. Even the number of processes in a basic systemd setup is higher than the total number of scripts in many SysV init systems.
I think my "mentality" fits perfectly in the Unix world, where a stable system is built using relatively small, specialized building blocks that do their stuff really well. Using Slackware, I can teach a group of Windows novices how to install and maintain a Linux system in a matter of hours. I know this, because I do just that once a year.

A systemd'ed Linux looks a lot more like Windows, and if I'm going to use something that looks like Windows, it may as well BE Windows. It's a lot easier to sell an upgrade to the latest version of SharePoint, Exchange or Microsoft SQL Server than it is to sell a migration to a Linux-based platform.

5 members found this post helpful.

I am not mocking, but you did not answer the question.

You said that you do not have any sysvinit machines, so if you don't even have a Slackware system running why are you so insistent posting to this thread in a Slackware forum?

However, I did use similar wording as your question for a reason.

Why not spend your efforts in the Red Hat or Gentoo forums? It seems to fit a lot more your "mentality" (no offense), than the Slackware world.

The reason was to hilight what strikes me as the childish immaturity of your posts in this thread (my personal opinion) and the absence of any useful content among all those words. It always comes down to...

"Does not!", "Does too!", "Is not!", "Is too!", "Will not!", "Will too!"...

So I thought I would see how you reacted to your own non-offensive empty suggestion for others being asked of yourself... and you did not seem to like it at all.

But you still did not answer the question...

If you don't even have a Slackware system running why are you so insistent posting to this thread in a Slackware forum?

+1 on that. I work as a professional Linux trainer for a series of companies in South France. I'm using Slackware for the job, with bits of Debian and CentOS.

@astrogeek, @Ser Olmy

Please, do not feed the troll. Thank you.