F11 Proposal: Stabilization
Les Mikesell
lesmikesell at gmail.com
Wed Nov 19 05:14:24 UTC 2008
Jesse Keating wrote:
> On Tue, 2008-11-18 at 20:17 -0600, Les Mikesell wrote:
>> Slightly different. More like updates and updates-tested. Something
>> you might opt-into after your machine was working the way you want and
>> you don't want to take unnecessary chances. But it might be even better
>> to have some sort of per-package risk rating that would go down with age
>> unless problems were reported and a per-client choice of how
>> bleeding-edge to go. And packages being pushed to fix security or
>> serious known problems could be added with a negative risk rating if the
>> packager is sure that it will make things better instead of worse.
>
> Given that we have a hard enough time keeping things straight and
> getting feedback for updates-testing, what makes you think we'll do a
> better job by adding a 3rd repo into the mix? And what are 3rd party
> repos supposed to target?
So what about the risk-factor rating concept? Suppose users had an
option that would ignore updates for about a week, then take them if
there were no new bugzilla reports, longer if there were, with
everything in one repo and some override options for special
circumstances? 3rd party repos could match the technique or not.
Updates or installs of new packages that would pull dependencies higher
than your acceptable risk factor could just fail.
--
Les Mikesell
lesmikesell at gmail.com
More information about the devel
mailing list