Todd Meade wrote:
I like the part about “the microkernel experiment has failed”, this is
sort of like a horse and buggy enthusiast (circa 1920) claiming that the
“horseless carriage” experiment has failed due to a lack of paved roads
(whereas the superior traction of the horse so greatly improves
performance on dirt roads).
Thanks for the link…
Rennie
Rennie Allen <rallen@csical.com> wrote:
Todd Meade wrote:
http://www.linuxjournal.com/article.php?sid=6105I like the part about “the microkernel experiment has failed”, this is
sort of like a horse and buggy enthusiast (circa 1920) claiming that the
“horseless carriage” experiment has failed due to a lack of paved roads
(whereas the superior traction of the horse so greatly improves
performance on dirt roads).
Yes, quite the troll.
Or the bits assuming you can’t recover from Fsys crashing. Or from
slaying Fsys. Or, that it will require a developper at the helm. Or,
even if you can’t fully recover, you can do a sane shutdown of other
pieces, and that may be incredibly important… returning a robot arm
to a safe position, dumping audit or transaction$ data accross a
network link, or other safe shutdown procedures. If your filesystem has
taken the kernel with it, you aren’t gonna be able to do that.
-David
QNX Training Services
http://www.qnx.com/support/training/
Please followup in this newsgroup if you have further questions.
<ad2sp5$hvq$1@inn.qnx.com>
How come this guy doesn’t mention WindowsNT?
It’s (um, well sort of) microkernel. (NT3.51 was rather pure microkernel)
Just we users didn’t benefit from that due to Win32 wrappings.
Theoretically you can kill and restart device drivers under NT too right?
kabe