The Hurd

[Frank Heckenbach]

Jeroen Dekkers wrote:

> > I also know that the Hurd is still under development. Extending
> > Marcus' usage of "dead", this means it's not born yet.
> 
> The Hurd has an increasing number of users. Does libc5 has that?

You still don't get my point. I don't claim that libc5 is better
than the Hurd (what a comparison ;-) or something like that. I just
don't like seeing free software advocates putting down other free
software with the same unobjective choice of words we're used to
hear from proprietary advocates (this applies to all sides in this
thread).

> > I'm no expert on this matter, but from what I've read I'm quite sure
> > that IPC is not the issue. It's about context switching etc.,
> > AFAIUI.
> 
> That can be pretty fast. On the i386 a context switch costs pretty
> much, but on other processors it's even less if I'm right. Because 10
> years ago microkernels were slow everybody thinks they are slow
> today. But they are fast today!

I'm sure when the Hurd will have been born, we'll soon see some
benchmarks, and if it's really as fast or faster than Tux, everybody
will switch, and there will be no market for Tux anymore, so it will
die quickly.

Frank

-- 
Frank Heckenbach, frank at g-n-u.de
http://fjf.gnu.de/
GnuPG and PGP keys: http://fjf.gnu.de/plan (7977168E)