[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Bits from the Release Team - Kicking off Wheezy



On 02/05/11 at 10:31 +0200, Martin Zobel-Helas wrote:
> Hi, 
> 
> On Sun May 01, 2011 at 21:53:58 +0200, Lucas Nussbaum wrote:
> > On 01/05/11 at 20:51 +0200, Martin Zobel-Helas wrote:
> > > Hi, 
> > > 
> > > On Sun May 01, 2011 at 20:02:51 +0200, Lucas Nussbaum wrote:
> > > > 2. determine who is in support of each action plan, through a GR or a
> > > > poll.
> > > 
> > > I don't think we need a GR for that. Those who are interested in rolling
> > > releases could show that they are interested and just doing so (like
> > > Norbert/formorer/Rhona/...  did with backports, like Joey Hess did with
> > > testing-security, like Andi and me did with volatile, ...).
> > > 
> > > I am aware this might need changes in some of Debian's infrastructure,
> > > but i am quite sure if you provide help/patches/... those will be
> > > implemented.
> > 
> > I don't see how that could work.
> > Iet's assume that the goal is to demonstrate the interest in the "rename
> > testing to rolling" scenario, without even talking about what to do during
> > freezes.
> > 
> > The first steps of the implementation will include:
> > - rename testing to rolling. I don't see how ftpmasters would do it
> >   without a consensus that this is something wanted by the project.
> > - communicate officially, to the general public, that rolling is not
> >   only a development branch, but also suited for use by the general
> >   public (given known limitations). I don't see how the press team would
> >   publish something like that without a consensus that it's what the
> >   project thinks.
> > 
> > What was applicable for backports, testing-security or volatile is not
> > applicable here, because the implications for the project are deeper.
> > It's not about adding a suite with some different packages in the
> > margin. It's about shifting developers' focus and user attention a bit.
> 
> No, it just needs that rolling is running on a different dak instance as
> testing. The same we had for volatile, the same we had for backports.
> The team (whoever that is) wo is interested in the rolling releases can
> show it is worth the effort, then we can start thinking integrating it
> back into the main archive.

What's the point?
Outside of freezes, rolling == testing. What's the point of running a
separate dak instance?
It would be about as efficient to collect statements of users saying "if
rolling existed, I would use it".

- Lucas


Reply to: