Re: Journalling filesystems

[Alfred M. Szmidt]

   >    I think they should be implemented in the Hurd.
   >
   > Would you like todo this?

   Yes, I would.

Then do so.

   But as Ognyan says: But we are far away from all that, and IMHO we
   should first finish porting to L4 before dealing with such stuff.

Where does he say this? I can't see such a message from him that
states this.

As for this "waiting philosophy", I consider it a disease that people
have around these parts.  If people wait and wait then nothing will
get done. When should you _stop_ waiting?  Take the L4 example, say a
better kernel pops up just when Hurd/L4 reaches the stability we have
now, should we say "oh, lets wait until this kernel is done because it
is so much better then L4, and then port the Hurd to that" (this was
the scenario we had while we waited for the L4 X2 specification!)?

Instead, just sit down and do the work instead.  Like write a
journaling file-system for the Hurd, it can always be ported when
things change.  And things will _always_ change, it is life after all.


So, shutup and code.