Jamie Lokier wrote:
> Xuan Baldauf wrote:
> > lol :o)
> >
> > The difference is: your ft script belongs to the distros, reiserfs to the
> > kernel. Ask some distros to include your script, maybe it is considered
> > useful. ;o) (Maybe reiserfs also belongs to distros, not to the kernel, but I
> > thought that we do not really want to go this way (and break linux into
> > distro-specific linux like latin into italian, french, spanish...) But maybe
> > I'm wrong...)
>
> Linux is already distro-specific. Different distributions already use
> different, customised kernels. You didn't know that?
I know that, but I do not like it very much. (Though I do not know wether splitting
OSs is somewhat evolutionary.) At least for the linux kernel "In a world without
walls and fences, who needs Windows and Gates." would be lesser true. If I would
want reiserfs only for people in Germany, for example, I would not need to comment
anything, because I think that it's pretty probable that SuSE will include reiserfs
in their kernel if it's not done.
> Some of them may even ship with Reiserfs.
I think so.
> Anyway, the arguments against including Reiserfs at this time are not
> about market positioning or user information. They are about technical
> development and maintenance issues. An fs in the kernel tree is not
> independent of the generic VFS code. Once in the tree, the two are
> evolved together. This holds even for "experimental" code.
>
> Good fs code of general interest generally does get into the kernel
> tree, if it's clean and works as the current VFS expects.
Mhh... it works, but it is deemed unclean and|or out of sync. Would some promise of
the reiserfs developers (I do not know wether they would do that, but I think so)
to clean and sync the code with the current VFS within the next (say 3) weeks help?
(I don't know how much time it would take) I think they will do it anyway, but
missing 2.4 and waiting another 80 weeks because being synced 2 weeks after a
negative inclusion decision is somewhat hard and disappointing, especially after
putting highest efforts for the 2.4 months. If reiserfs is decided to have missed
2.4, development will slow down, because reiserfs developers would have to maintain
2 versions (for 2.4 and 2.5), because there would be less user interaction (feature
requests, bug reports, maybe co-developers) and because people would see their work
and time spent the last months between their bills in the bin and not in /sbin. ;o)
>
>
> > > > Compare this to some dictatures... say Serbia. Most people have the
> > > > chance to change the political system (they can try to form opposition
> > > > and demonstrate, etc.), but they do not have the choice, because they
> > > > have no democratic elections.
> ...
> > > If that's what chance and choice mean, then users _do_ have the choice
> > > to install Reiserfs.
> >
> > I do not get that point (maybe I'm to tired). Is the user asked "reiserfs"
> > within "make menuconfig" after freshly installing a current kernel? Until the
> > answer to this question is "no", users have only the chance, but not the
> > choice...
>
> Nonsense. With your political dictatorship example you have defined
> choice to mean "if they want to, they can". And chance to mean "they
> can try but they won't necessarily get it".
No, I defined choice to mean "you are asked for" (this does not necessarily mean
that you get it, that's another question), and I define chance to mean "if you want
to AND you know that it's possible, you can"
> Actually a democracy doesn't give people (plural not "the people") the
> choice to change their political system -- tyranny of the majority may
> prevent that or, often, even if the majority wish to change something
> that may not be possible within the chosen democratic system. (I know
> of no perfect system).
I know that. And I also do not know a perfect system, as I did not know a perfect
example for representing chance and choice and the difference between them. Maybe I
should add to the example that I mean "the chance to influence" and "the choice to
influence" (though I do not know wether this is correct in english grammar...)
Another definition: chance: you have to be active to get the results; choice: your
required activity is reduced to say "yes" or "no".
Hope this untangles. :o)
> enjoy,
> -- Jamie
Xuân. :o)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Mar 15 2000 - 21:00:25 EST