On the conviction of Hans Reiser

Oh, I didn't mean specifically about ReiserFS; I can't really comment on whether or not
ReiserFS was really a good idea or not. But much of the response to it on LKML struck me at
the time as 'POSIX was good enough for the late 60s, so it is good enough now, and you'll
implement it or shut up' rather than a reasoned response to actual flaws in Reiser. It is that
kind of uncritical 'this is how we've always done it' thinking that I can't stand, and which,
frankly, I think determines a little too much of how most of us (proprietary or open) develop
software.