Systemd Development Skyrocketed This Year

[alaviss]

I think most people like the idea, but have problems with the implementation.

I dont like to find 4GB coredump files of program crashes on my sys partition for example.

Your apps crash that much? Anyway, you could turn off coredump by modifying /etc/systemd/coredump.conf

[edmon]

Your apps crash that much? Anyway, you could turn off coredump by modifying /etc/systemd/coredump.conf

you can do this, you can do that?
And why?
Did i have to turn on/off some 'feature/configuration/daemon' in systemd before it wipe my hdd or install botnet and start to flood CIA???

[gilboa]

you can do this, you can do that?
... or install botnet and start to flood CIA???

Troll much?

[MoonMoon]

For what I can see in the graphics...Kay Sievers is the big guy how hold the credit's of SystemD!

Maybe Lennart is not so guilty like we thought...

The only one guilty are people demanding software that is written and works the the way they want without doing anything to contribute that. Blame people for not maintaining Consolekit if you want, blame people for not coming up with alternatives to what systemd provides if you want, but why would you blame people that actually are doing something other developers want?

[duby229]

The only one guilty are people demanding software that is written and works the the way they want without doing anything to contribute that. Blame people for not maintaining Consolekit if you want, blame people for not coming up with alternatives to what systemd provides if you want, but why would you blame people that actually are doing something other developers want?

The problem is that systemd isnt providing anything.... All they do is suck up existing projects that they know will screw the maximum number of people. The more this goes on, the more duplication of work will need to happen. Projects that existed long before systemd, now part of systemd, needs to be replaced with new standalone projects. Those standalone replacements don't exist yet, so systemd is the only real choice for too many distros

[phatfish]

I really hope there is some hidden code in systemd which detects when it is installed on every Linux system in existence. It will then encrypt all copies of the source code with a key only Lennart knows and he will proceed to sell the source to Microsoft who will continue development.

Maybe when all the trolls are proved right they will finally shut up.

[duby229]

I really hope there is some hidden code in systemd which detects when it is installed on every Linux system in existence. It will then encrypt all copies of the source code with a key only Lennart knows and he will proceed to sell the source to Microsoft who will continue development.

Maybe when all the trolls are proved right they will finally shut up.

Nobody here said anything like that but you. Obviously that's not what's happening anyway.

[ciplogic]

The problem is that systemd isnt providing anything.... All they do is suck up existing projects that they know will screw the maximum number of people. The more this goes on, the more duplication of work will need to happen. Projects that existed long before systemd, now part of systemd, needs to be replaced with new standalone projects. Those standalone replacements don't exist yet, so systemd is the only real choice for too many distros

Did you read what systemd does? Then, it looks to me that it provides.

What this project does, between other things (from memory):
- it gives an (more) unified way to maintain services startup
- it gives a clean way to restart crashing services
- it likely improves the startup time on most machines
- it is more modern and more maintained than init and it has the biggest chance to bring one part of userspace to be less hacky that it ever was

It is in a way like Mir or Wayland: they improve and address long standing issues in X. If there is enough support, no one should complain as you don't do the work to fix a PulseAudio driver bug, but you would likely complain on here for Pulse being "crappy" or "sucking resources".

I am not a fan of Lennard but I have to say: "thank you" as for all other projects that at least give the a real improvement in opensource projects. People complaing are easy to find, but work into projects who everyone can use, are much harder to be found, isn't it?

[duby229]

Did you read what systemd does? Then, it looks to me that it provides.

What this project does, between other things (from memory):
- it gives an (more) unified way to maintain services startup
- it gives a clean way to restart crashing services
- it likely improves the startup time on most machines
- it is more modern and more maintained than init and it has the biggest chance to bring one part of userspace to be less hacky that it ever was

It is in a way like Mir or Wayland: they improve and address long standing issues in X. If there is enough support, no one should complain as you don't do the work to fix a PulseAudio driver bug, but you would likely complain on here for Pulse being "crappy" or "sucking resources".

I am not a fan of Lennard but I have to say: "thank you" as for all other projects that at least give the a real improvement in opensource projects. People complaing are easy to find, but work into projects who everyone can use, are much harder to be found, isn't it?

I know I have long standing issues that you are not be aware of, so in a way I apologize for that. I know that systemd serves vital functionality. A lot of it. In my opinion too much of it.

Here's basically what I think of systemd it's like a spinning plasmoid that sucks up every niche it can find until all that's left is this gigantic fiery ball of junk.

[DMJC]

I'm still struggling to see where systemd makes my experience better? I don't care about boot times, that's what SSDs and laptops with sleep/resume solved. I don't care about the logging/service modifications since all those things already worked. I don't care about projects like consolekit etc being abandoned since they're likely to be abandoned within systemd anyway. (No one works on unsexy code whether it's in a big project or a standalone one because *surprise* it's boring, Wine is a great example of this). I've had a computer left completely unbootable due to policykit breaking on debian which required a reinstall (I sought help on IRC and noone in #systemd actually provided any help. Aside from telling me to view a log file which didn't give me any clues where the break was.) Avahi/pulseaudio/wifi/graphics all work the same as they used to and on a laptop that's about the only space outside of desktop apps where you can improve. I still can't manage openchange using Exchange Server Management Tools or have C/C++ desktop tools for managing Samba4/OpenChange so there's no big changes happening on the work side. I still don't have a working clone of Caligari Truespace on LInux or a decent 3d ship editor for VegaStrike (although those last two I'm working on myself ) I'm not too fussed either way with systemd, it seems to be the happening thing, just don't break GNUstep or my game stuff and I won't be too angry about it, but I still don't see the benefits to systemd. If someone has an argument other than boot times and maintaining abandoned projects, I'd be interested in hearing it.