Eelco Dolstra's Ph.D. thesis describing Nix, "The Purely Functional Software Deployment Model": https://nixos.org/~eelco/pubs/phd-thesis.pdf
Armijn Hemel's master's thesis describing NixOS: https://nixos.org/docs/SCR-2005-091.pdf
Or this more recent journal article about Nix/NixOS which is a good introduction: https://nixos.org/~eelco/pubs/nixos-jfp-final.pdf ("NixOS: A Purely Functional Linux Distribution").
The Open Group would disagree.
Personally, I'd call OS X unix-like but not unix-alike, which perhaps sounds like hair splitting but having admined multiple unices there's definitely a difference, even if I can't figure out how exactly to articulate it beyond "it doesn't feel the same".
[1] https://developer.apple.com/library/mac/documentation/MacOSX...
I'm thinking perhaps some measure of porting effort is the appropriate metric. There is a reason Mac users can't simply point to a bsdports repo.
So dll dependencies are crazy-making, and Nix is attacking that. traditional init is getting creaky, launchd and systemd are attacking that.
With respect to systemd, I love how it has motivated people to try to come up with something better than both traditional init and systemd.
NixOS is rebuilding important pieces of the foundation. Why? To address a basket of problems we have been complaining about with Unix for a very long time. Is it worth the hassle? Well, have you ever lived through a kitchen remodel in your house? That's a hassle, too, but you end up with a better kitchen.
IMHO NixOS is pointing the way forward, but is probably not the last word in next-generation operating systems. The existence of NixOS and Guix (and similar) are exciting, because they are part of the ferment creating the thing on which we will eventually converge.