unix_fan <tmel...@web.de> writes:
> Is mach dead?
Well. Mostly. Mach was a huge research project over many years and
places. A lot was experimented with it and a lot learned from that.
(Also we now know how not to do certain things...)
The latest publically visible occurences of Mach were within MkLinux
(Linux on top of Mach), Mac OS X (somehow uses Mach) and the GNU/Hurd
system, which also is still using Mach as its underlying microkernel.
Regards,
Thomas
No. And now try trolling in another group.
Martin
--
"For the Snark's a peculiar creature, that won't
Be caught in a commonplace way.
Do all that you know, and try all that you don't;
Not a chance must be wasted to-day!"
Martin Etteldorf <ette...@email.lu> writes:
> unix_fan <tmel...@web.de> wrote:
>> Is mach dead?
>
> No.
Where is Mach really alive? (Seriously.)
Regards,
Thomas
In MachTen, Mac OS X and Tru64.
> unix_fan <tmel...@web.de> wrote:
>> Is mach dead?
>
> No. And now try trolling in another group.
>
>
> Martin
Oh, okay. Mach is not dead, but you want to make damn sure that your
newsgroup remains so, right?
> Hello!
>
> unix_fan <tmel...@web.de> writes:
>> Is mach dead?
>
> Well. Mostly. Mach was a huge research project over many years and
> places. A lot was experimented with it and a lot learned from that.
> (Also we now know how not to do certain things...)
Oh, that's intriquing! Can you give any examples?
>
> The latest publically visible occurences of Mach were within MkLinux
> (Linux on top of Mach), Mac OS X (somehow uses Mach) and the GNU/Hurd
> system, which also is still using Mach as its underlying microkernel.
Well, Mac OS X is apparently viable. Somebody mentioned Tru64 Unix. I've
used that and didn't know it was mach-based ... it almost seems incredible
to me - it's a fundamentally different approach. If it were better, more
platforms would use it. If it were inferior, a company like Digital would
have been able to jetison it quickly.
Can it be that mach is in Tru64 and Mac OS X in name only?
Is the idea of a message-passing kernel - with user-space conventional
kernel functionality - dead?