> see if one day a replacement for systemd comes along
Part of the critique of systemd is the basic architectural choice of having this monolithic layer between regular user apps and the kernel. So, in a sense, the idea is _not_ to replace systemd with a better-written systemd, but to do things differently.
Yes but in doing things differently with the new one people will defend systemd the same way people defend sysvinit was my point. They will claim the new way is too complicated or tries to replace too much.
I might not be getting the point of the talk but I really appreciate the argument that Benno Rice presents.
Part of the critique of systemd is the basic architectural choice of having this monolithic layer between regular user apps and the kernel. So, in a sense, the idea is _not_ to replace systemd with a better-written systemd, but to do things differently.