On Tue, May 4, 2010 at 2:57 PM, ron minnich <rminnich@gmail.com> wrote:

> On Tue, May 4, 2010 at 2:23 AM, hiro <23hiro@googlemail.com> wrote:
> > The more interesting question is: who doesn't agree, and why?
> >
> > On 5/4/10, Pavel Klinkovsky <pavel.klinkovsky@gmail.com> wrote:
> >>> maybe it is time to try to pack-port some of Erik's stuff to the
> canonical
> >>> source.
> >> I fully agree.
> >>
> >> Pavel
>
> The argument here is that we need to sync 9atom back to sources, and
> that may well be true.
>
> But there's another path:
> use mercurial to create a clone of
> http://bitbucket.org/rminnich/sysfromiso
> you can call it 9atom.
>
> You can put your changes there.
>
> Then you can use the mercurial tools to continually refresh your 9atom
> tree from sysfromiso.
>
> In that way, you can provide a 9atom tree that is perfectly in sync
> with sources, and it is easy for others to see what you have done.
> And, most importantly, the maintainers of the main tree can easily see
> what they need to see, and figure out what ought to come back to the
> mainline, and pull back things that make sense to pull back.
>
> There is real precedent nowadays for people to maintain forks of a
> kernel tree, where they can experiment, and do so in a way that makes
> merges back to the mainline easy.
>
> ron
>
> Right, forks aren't always evil and to be avoided.  In some cases they're
just perfect for organized experimentation.