What is the problem with systemd?

Hey guys, i've been seeing a lot of hate of the internet on how "bloated" systemd is, and that it doesn't follow the UNIX way, what are you guys opinion on this?

Systemd is a benefit in my opinion. "Bloated" is not the same as "more features", and the "more features" of systemd makes system diagnostics A LOT easier and faster.

I do understand all the commotion though. Debian sysadmins are used to never having to learn anything new, because Debian is so conservative. That's also why I use Debian on some servers, because it's very low maintenance, both on the software update side AND the sysadmin knowledge side. But the world moves on, and for some server applications, that are not as passive as simple web servers or postfix systems, a high degree of system diagnostics and fine control of daemons is required, and systemd offers just that, and in my opinion, it's a very welcome enhancement, and because of the years that it's been proven in RPM-distros and Arch/Gentoo based distros, it's definitely compatible with the Debian philosophy of ultimate stability.

One of the features of systemd is that all systemd CLI commands are also automatically aliased, so when you use the SysV init command, it's automatically translated into the corresponding systemd command, so users that don't want to use the extra features of systemd, don't have to learn anything new.

On a deeper level, a lot of the controversy has to do with "FLOSS activism" and with inter-personal friction. Systemd was mainly developed by Lennart Poettering and Kai Sievers, and there has been a lot of criticism lately against Kai Sievers, for a "capital sin" in the FOSS-world, which is to shift problems to other devs by just coding ahead and breaking compatibility with pre-existing or upstream packages and also modifying code of other projects to make it fit into one's own project. That is a problem in the quality management system for which open source software is known, and it has even been dealt with by Linux Torvalds himself. As far as the influence of FLOSS activism goes, many hardcore FLOSSers object to the merger of udev into systemd, or the fact that it integrates DBus. They also object to the binary Systemd logs, and prefer syslog event logging, for a number of reasons, which in my opinion boil down to the fact that they have to adapt their usual workflow when switching to systemd, because the systemd logging system does have some pretty big benefits.

Especially users that develop a lot on systems, usually like systemd, because one of the features of systemd is that it allows an admin to snapshot a complete system state, and that may save a lot of time.

I see... nice read as always, Zoltan, thank you