A Glorious Vision of Our Shared Update Feedback Future (bodhi, karma, and proventesters, oh my)

[Adam Williamson]

On Tue, 2011-11-22 at 22:17 +0100, Sandro "red" Mathys wrote:
> On Tue, Nov 22, 2011 at 22:03, Adam Williamson <awilliam at redhat.com> wrote:
> > 2. Any update marked as 'critpath breaking' by a proven tester would be
> > blocked from being pushed stable at all - automatically or manually -
> > until the PT modified the feedback or it was overridden by someone with
> > appropriately godlike powers (TBD, but probably not just the maintainer)
> 
> Here I'd add something like "three proventesters can turn over one
> proventester" and only if that doesn't work activate someone with
> godlike powers. Just because godlike powers should never have to be
> used in an ideal community/world and I think this overturn rule can
> help to get us a little closer to ideal ;)

Possibly, but in that case, I'd want the mechanism to require the
'overturning' PTs to explicitly consider the 'omg raptors' PT. i.e., if
3 PTs check 'doesn't break critpath' and then one PT checks 'breaks
critpath!', those three earlier tests should not be considered to
override the one later one. Other PTs would have to explicitly say that
they tested whatever the 'omg raptors!' PT said was broken and could not
reproduce, in order to overturn their feedback.

(And then we'd go after the incorrect PT with the 2x4 with a rusty nail
in it...)
-- 
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | identi.ca: adamwfedora
http://www.happyassassin.net