From the article -- Windows NT had a strong service model from the beginning, he said, and Mac OS has one now in the form of launchd. Other systems had to play a catch-up game to get there.
Reminds me of the saying, "and if your friends all jumped off a cliff, would you jump off too?" which exasperated parents use to council their children that emulating others needs to be done responsibly.
There is a big difference here. To continue with the cliff analogy: Launchd ran fast enough to clear the cliff and was rewarded by a refreshing dive into the cool water; it was implemented in a competent manner and consequently didn't cause people to become upset. Systemd on the other-hand can't run without stumbling, it hobbled off the cliff and bounced off every rock on the way down. It made lots of people pissed off because it interfered with their routines.
If systemd didn't cause problems for people, virtually no user would even know what it was, and consequently, virtually no user would hate it. If you ask your average OSX user what launchd is, they'll come up with a blank because launchd isn't causing trouble for them. If you ask your average linux desktop user what systemd is, a large portion of them will recognize the name, and many of them will chew your ear off out of frustration.
At the end of the day it's not about "adherence to philosophies" or any similarly dubious concepts. Whether or not software is hated is ultimately determined by whether or not it causes problems for people.
That's not the problem. Windows NT did have that right from the beginning. The problem is the second sentence, because it ignores that so too did Unix from the same timeframe. AT&T System 5 Release 4 had the Service Access Facility from 1988. AIX had the System Resource Controller from version 3.2. Other systems were not playing catch-up. They had this stuff too, at around the same time.
Benno Rice's explanation is actually ahistorical here. What actually happened is that the world spent a lot of time "cloning" existing Unix softwares during that time period, for reasons that we all known, and a lot of the time the clones were behind the times and did not catch up. By the time that Miquel van Smoorenburg cloned AT&T Unix init and rc for Minix, for example, what xe was cloning had already become years out of date.
Reminds me of the saying, "and if your friends all jumped off a cliff, would you jump off too?" which exasperated parents use to council their children that emulating others needs to be done responsibly.