Re: Building against testing [was Re: "Recompile with libc6 from testing" <-- doesn't help i think!]
On Mon, 16 Sep 2002, Anthony Towns wrote:
> On Sun, Sep 15, 2002 at 11:17:51PM -0500, Adam Heath wrote:
> > On Mon, 16 Sep 2002, Anthony Towns wrote:
> > > Been considering whether maybe having some sort of automated "hey, you've
> > > made a dozen debconf uploads recently, none of them have had RC bugs,
> > > so I'm going to divert any more uploads you make to experimental for
> > > the next 11 days, so the current version can get into testing" might be
> > > convenient, though.
> > Murphy's law would hold, and the last upload you consider would have an RC
> > bug.
> That wouldn't be particularly catastrophic though. If it's found after,
> say, five days, you just move the latest upload that got diverted to
> experimental back to unstable, and keep going.
> Alternatively, you could do something like have uploads with an urgency
> of "risky" or something get diverted to experimental until the previous
> version either (a) makes it to testing, or (b) spends, say, fourteen
> days not getting it to testing and "times out".
Hmm. I really really like this idea. What would it take to make it happen?