The next step of GNU in pure technical perspective (was: When can we expect a version 1.0 of the GNU Operating System?)

[Nala Ginrut]

On Wed, 2014-12-03 at 02:36 +0100, Olaf Buddenhagen wrote:
> Hi,
> 
> On Fri, Nov 21, 2014 at 02:43:46PM +0100, Garreau, Alexandre wrote:
> 
> > I heard that on #hurdfr I think (the ???nightmare??? of pipes and the
> > fact GNU Hurd would be forced to follow POSIX, also that I remember
> > having seen braurn doing a minifork of Hurd to experiment features
> > without following POSIX, I don???t remember the name but it looked
> > like a proper name???),
> 
> "x15" is the name he (Richard Braun) uses for his Hurd-related kernel
> experiments. Note though that the focus of his work has shifted several
> times over the past years -- so "x15" can actually refer to a couple
> somewhat different things...
> 

Actually, x15 is an alternative of Hurd rather than a Hurd-related
kernel, of course, experimental. I thought Braun has dropped this
project few months ago, but I've confirmed this morning, it's updating
when he's free. ;-)
I don't know the purpose of its design yet. But I think the current Hurd
is hard to change its core (GNU Mach) to others without large changes,
say, L4. I guess x15 could be a better design, dunno.

IMO, GNU as an OS has great flexibility to change its kernel as will.
From pure technical perspective, it's very good design if someday we can
prove it in industrial way. For now, it's not persuasive yet.

Maybe the next step of GNU is not to provide a workable OS anymore,
since we've made it, and it became an important part of the industry.
According to its design, kernel-free could be a significant feature of
GNU, to provide further freedom in technical way. Besides, maybe
language-free could be another one, if multi-lang in Guile platform
could be finished in the future. 

kernel/language-free here means users don't have to worry about what
kernel/language should they choose, just use whatever they like, or
choose it according to the design of their projects.

This sounds hard, and it's very hard actually.
If we think about this point, the purpose of Hurd and Guile become
meaningful in practice, not only "oh it's just interesting, so we
continue".
So it's worth to consider, IMO. ;-)

Comments?


-- 
GNU Powered it
GPL Protected it
GOD Blessed it

HFG - NalaGinrut

Fingerprint 77F8 7D5B B425 E505 535E  C365 9182 7E28 EE78 E925

signature.asc
Description: This is a digitally signed message part