The email from Linus Torvalds to Mauro Carvalho illustrates two things:

1. Torvalds' somewhat abusive style
2. Torvalds' firm no-regression policy: user code that used to work should never be broken by a kernel update.

How do these factors play out in the success of Linux?

A third factor is Torvalds' policy on contributions: all are welcome, but they start out as non-mainline extensions, patches or modules.

Start with the SCO Wars

Wednesday March 1: Linux History, "Containerization"

Open-source security