Subject: Re: Dual 500 G4
From: Timothy A. Seufert (tas@mindspring.com)
Date: Sat Feb 17 2001 - 05:56:13 MST
At 2:19 PM -0800 2/16/01, Brandyn Webb wrote:
> By then, though, OS-X will be out. Is there any
>consistent sentiment for how Linux/mol will fare after that?
I for one plan on using both.
>My personal attraction to OS-X is simply a preference for
>microkernel vs. monolithic,
Pet peeve: OS X is not a microkernel. Really. It's monolithic,
though in some regards it is structured like a microkernel.
True microkernel operating systems are no longer in vogue, because
they have performance problems.
What really matters is clean OS design. One possible path to a clean
OS design is to use some of the principles developed for
microkernels. But it is not the only path, and there is no reason to
treat microkernelishness as a Holy Grail.
>plus industrial-strength support
>for hardware (DVD-R, etc..). Here's a question: How hard
>would it be to make Darwin run Linux (ppc) binaries, or to
>run mol? I just wonder if a Mach kernel with gnu trappings
>wouldn't be a more responsive, more reliable, compatible
>linux-equivalent (something like Hurd is trying to be)...
Very doubtful that it would be any faster overall. Linux is already
quite fast.
Hurd is (IMO) mainly trying to survive in the face of overwhelming
odds against it. It's not exactly a model of efficiency, and part of
the reason for that is that it was designed as a true microkernel OS.
Tim Seufert
This archive was generated by hypermail 2a24 : Sat Feb 17 2001 - 05:57:21 MST