On Wed, 2003-05-21 at 10:47, Aryan Ameri wrote: > On Wednesday 21 May 2003 15:53, Kent West wrote: > > Paul Johnson wrote: > > >-----BEGIN PGP SIGNED MESSAGE----- > > >Hash: SHA1 > > > > > >On Tue, May 20, 2003 at 09:57:05PM +0200, David Fokkema wrote: > > >>IIRC, GNU will change to the hurd sometime in the future. This does > > >> not matter for debian, of course, ;-) > > > > > >No, but I have a feeling Debian will get *lots* of users, since > > > we're still the first and only Hurd distro... > > > > At my last LUG meeting, the HURD was mentioned, and there was quite a > > bit of interest in playing with it. I've been reading up on it this > > week, and plan to install it on a box very shortly. Although from > > what I've read, I'm anxious to get to the L4 kernel (fast) and away > > from the Mach kernel (slow). I'm pretty sure that Mac OS/X uses the > > Mach kernel (but with a monolithic "kernel" on top instead of many > > servers the way that the Hurd, and only the Hurd, uses), so I guess > > the Mach kernel isn't _too_ slow. > > > Considering that RMS promotes GNU free software applications over their > non-GNU free software equivalents (Gnome versus KDE comes to mind), it > will be interesting to see the GNU project and FSF part ways with > Linux. > > >From a technology point of view, Hurd with the idea of a true > Microkernel, is exciting, . But it certainly is lacking interest these > days, and is still in the alpha stage. Nowadays hundereds of developers > are either individually, or on behalf of companies like SuSE, RedHat, > SGI, IBM, HP ... are working on the kernel. I can't see Hurd catching > up with Linux, considering that the Linux kernel is progressing with > such a rapid pace. If things continue like this, then IMHO Linux 2.8 > might even catch up with Solaris. > > As much as I like RMS and his ideas, I beleive that when in 1995-96 it > became apparent that Linux was portable, FSF should have stopped the > development of Hurd. > > Just my $0.02 > > Cheers > -- > /* There is SCO owned IP all over the Linux kernel. SCO will hunt them. > Free software infidels are liars. We will kill them all, and roast their > stomach in hell. Our estimates show that all slashodot viewers will die. > --Mohammad Al-Sahhaf SCO Sopkesman, Former Iraqi information minister*/ > > Aryan Ameri I would suggest that while the last observation may be harsh (stopping the development of Hurd,) given the nature of free software it would also be unfeasible, as the handful(s) of interest in it conceptually would still be predisposed to continue developing it. I think there is potential in it as a concept and research effort, but I increasingly suspect that we are going to see a fork of Linux to extract a microkernel and servers from its codebase (while still abi compatible with traditional Linux) before the current limited resources available to the Hurd community get to a generally viable o/s. FSF connection to the Hurd reminds me of some old software attitudes of large tech corporations - the old "Not Built Here" resistance that honestly flies in the face of the Bazaar aspect of free software development. Shame that the intellectual value of free software still falls prey to license restrictions and even without the Cathedrals, still faces religious wars :( -- Mark L. Kahnt, FLMI/M, ALHC, HIA, AIAA, ACS, MHP ML Kahnt New Markets Consulting Tel: (613) 531-8684 / (613) 539-0935 Email: kahnt@hosehead.dyndns.org
Attachment:
signature.asc
Description: This is a digitally signed message part