On Mon, May 14, 2012 at 02:32:17PM +0000, Balwinder S Dheeman wrote:
> On 05/14/2012 05:00 PM, ***@polynum.com wrote:
> >
> > Hence, Plan9 is in part, by design, insulated from entropy.
>
> Plan 9 has never approached Unix in popularity, and has been primarily a
> research tool:
>
> Plan 9 failed simply because it fell short of being a
> compelling enough improvement on Unix to displace its ancestor.
> Compared to Plan 9, Unix creaks and clanks and has obvious rust
> spots, but it gets the job done well enough to hold its
> position. There is a lesson here for ambitious system
> architects: the most dangerous enemy of a better solution is an
> existing codebase that is just good enough. ? Eric S. Raymond[3]
>
I'm aware of this (and of who...). But this doesn't contradict what
I wrote: there are small "systems" out there, used by small
communities, that "evolved" from a reliable small codebase to huge
beasts, with no improvment made, but only an indefinite amount of
"tries" added, loosing the needle in a haystack.
Since Plan9 is small and the very spirit is to have, on the user
level, one small tool that does the job and no overlapping (a
mathematical partition) it is insulated from userland improvment---that
go to contrib. And since, on the kernel level, the principles are
few, before trying to adapt to a corner case taking the presence
of such hacks elsewhere as an excuse to add some more, you have to
dive in the whole because even a small piece has impact everywhere.
So even when there are short comings, the alternative solution is
never a panacea and one finally conclude that the original compromise
was a good one if not the best.
And seeing how Unices are fighting to try to get things working in an
environment not made for it (union fs for example; X having put the
network at the wrong articulation point and now trying to put back the
servers in the kernels etc.), the "technical merits" have more to do
with human inertia than with technicality. Inertia does exist; but "good
enough" is more: evolving exceeds my will to work and the material
benefits I can expect, and I will loose my position as a "Y system
wizard".
--
Thierry Laronde <tlaronde +AT+ polynum +dot+ com>
http://www.kergis.com/
Key fingerprint = 0FF7 E906 FBAF FE95 FD89 250D 52B1 AE95 6006 F40C