I read this article a while ago and really enjoyed it, despite my
personal appreciation of Linux (I use it almost exclusively at the
moment). I think his other points about computer science entering a
kind of creative plateau and suggesting biology as a new horizon for
trail blazing are fresh views, too.
Try to put yourself in Ken's shoes. He can and has written OSes from
scratch without the assistance of an example to clone, and at first
without a large extended community to bugfix the results. Also, he may
be coming from a school of developers who are happier working in
small, fully-funded, tightly-focused groups where design quality is
far more important than mere functionality of code. I've met a few
programmers like that, and I can understand why they'd feel that way.
It gives them a lot of creative control combined with the benefit of
being financially supported. That sort of collaboration looks like a
lot of fun to me, despite my love of the open source software concept.
Also, Ken's motivation for trying the latest version of Linux even on
a yearly basis is probably very low. From personal experience, I think
he's now wrong about Linux being less reliable than Microsoft
operating systems, but the term "reliability" must be carefully
defined before you can toss it around in any case.
Since I don't know that much about Ken, I should stop trying to
interpret his point of view!
I'm grateful for Linux and it does everything I need for the moment,
but I'd love to try running Plan9 (http://plan9.bell-labs.com/plan9/)
on my home network for a while :)