On Wed, May 14, 2008 at 01:12:00PM -0700, [EMAIL PROTECTED] wrote:

I'm not an expert but I think the counterargument to the "easier to understand"
is that you must implement a very complex asynchronous message passing
mechanism for your "simple" microkernel.  That's why RMS said Hurd took a long
time.

HURD lost it's momentum when Linux came out.  Suddenly there was no need
for it.  Whereas before people might want to run it, just because it was
the only free thing there was.  Once Linux came on the scene, followed by
various BSDs, HURD had to do something quite different to be useful.  It
really isn't much more than just something people poke at, especially since
it lags so far behind in HW support.

It's also going to be slow.  Whatever one might feel about microkernels,
HURD hasn't had a long period of tuning, so probably won't be very fast.

Also, Mach is far from small or simple.

David


--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list

Reply via email to