Thank Charlie.  But I just threw up after I read it. 

Sadly, UNIX's "prime directive" was to "keep it simple."  Or, as someone else describes it, create "small tools that did one job well." On the PDP-11, the lack of address space somewhat enforced this. With the 32-bit vax, we see cat -v and the like. I think "frameworks" are just a modern term for IBM's "access methods" of the 1960s. John Lions observed that the entire documentation set for UNIX V6 could be kept in a 3-ring binder, and, as his book showed, given the size, anyone could understand all of the kernel and the core systems ideas.

FWIW, Linux is not the first to fail. Years ago, I pointed out to Dennis that the System V Release 3 bootloader for the 3B was larger than the entire V6 kernel. I  have not looked at the size of systemd, but do you want to bet that it fails the same test?

But I digress. Someone (Henry Spencer, maybe) once said, "Good Taste is subjective. I have it, and you don't seem to." 

IMO systemd, was >>not<< a net positive - it falls so many of these tests WRT to good programming and good ideas.

Sigh ...

Clem