Re: hurd does NOT need /hurd
On Mon, May 20, 2002 at 04:13:22AM +0200, Rune B. Broberg wrote:
> Or if the Hurd forks, I presume
Interesting thought, although I am not exactly sure what this would mean.
The Hurd consists of individual cooperating servers. The natural way to
do what you want in the Hurd is just to add your servers to the system,
and run them. If you write replacements for the standard servers, there
is no reason they could still be considered part of the Hurd system and
put into /hurd, even if they directly compete with the standard servers,
and even if they are based on the same code.
A fork somehow implies that you have to choose between the one or the other.
For example, fork the Linux kernel and everyone must decide which version
he installs and uses. In the Hurd, any user could make this decision
himself (just as you can install emacs and xemacs and decide yourself, and
they are both in /bin, or one is in /bin and one in ~/bin).
There are plenty of RPC id numbers free, so if "forkers" want to have some
they can get them, I guess, if they have to introduce new subsystems (which
we want to avoid, because more RPCs means that cooperation becomes harder).
`Rhubarb is no Egyptian god.' Debian http://www.debian.org firstname.lastname@example.org
Marcus Brinkmann GNU http://www.gnu.org email@example.com
To UNSUBSCRIBE, email to firstname.lastname@example.org
with a subject of "unsubscribe". Trouble? Contact email@example.com