Re: Bug#798476: Returning to the requirement that Uploaders: contain humans
Am 2. August 2017 23:48:15 MESZ schrieb Sean Whitton <email@example.com>:
>Here is an updated diff for this bug, against the docbook version of
>the policy manual.
>I've also included a purely informative change which emphasises that
>packages that are team maintained in name only should be orphaned
>properly, with their maintainer field set to the QA team. This is
>already current best practice, but it's worth emphasising, because one
>might fail to orphan a package on the grounds that "someone else on the
>team might fix it", which is not true of a lot of teams.
>This purely informative change came out of a discussion at DebCamp with
>h01ger, gregoa and David Bremner. We are CCing -devel because we want
>to determine if there remains, in 2017, a consensus that we should not
>drop this requirement. We think that recent objections in the bug are
>about implementation details, rather than a concern to retain humans in
>the uploaders field.
>diff --git a/policy.xml b/policy.xml
>index 3daa532..4731507 100644
>@@ -1128,13 +1128,6 @@
> described in <xref linkend="s-f-Maintainer"/>.
>- If the maintainer of the package is a team of people with a
>- email address, the <literal>Uploaders</literal> control field
>- be present and must contain at least one human with their
>- email address. See <xref linkend="s-f-Uploaders"/> for the
>- of that field.
> An orphaned package is one with no current maintainer. Orphaned
> packages should have their <literal>Maintainer</literal> control
> field set to <literal>Debian QA Group
>@@ -1149,6 +1142,12 @@
>+ This includes packages with a group of people or team in the
>+ <literal>Maintainer</literal> control field. They should be
>+ orphaned if the team is not actively maintaining the package.
> <section id="s-descriptions">
>@@ -3448,13 +3447,6 @@ endif</programlisting>
> Maintainer field, and multiple entries must be comma separated.
>- This is normally an optional field, but if the
>- <literal>Maintainer</literal> control field names a group of
>- people and a shared email address, the
>- <literal>Uploaders</literal> field must be present and must
>- contain at least one human with their personal email
> The Uploaders field in <filename>debian/control</filename> can
> be folded.
(Please appologize the brevity, I don't have the time needed to word that
Well, I still think that not having a human explicitly named as in charge of
the package is not good and will cause more damage than it will bring
(Disclaimer: My view is biased with my actitivies in the MIA team)
Some remarks / questions I do not see adressed:
- If you have not a name on some task human nature tends toward that nonone
feels responsible at all. Like the German (fun) expansion of TEAM: Toll, Ein
Anderer Machts (Great, someone else it taking care)
- MANY team homepages are not updated or do not even exist. I think before
droping the requirement to have human uploaders this needs to be fixed by
policy and it must be RC(?) bug if the team homepage is
outdated/absent/inaccurate. The infomation about the teams *must* be in a way
that it can be easily found/parsed.
- There is currently no way to map a person to a team or to map a team to a
list of members -- if you need accurary. This is especially true for teams
that are smaller. - - When someon is going e.g MIA, we need to know which
teams are involved to trigger them.
- Not in all teams it is working tha everyone is looking at every package.
During the lipng transistion I filed many bugs on team-managed packages which
were never answered
- We have already the problem about "partially MIA" -- people holding several
pacakgages but neglecting several of them. Currently we have no real measures
of taking care of the neglected one, MIA team wise. This will be amplified
when there is no human responsible person named.
Diese Nachricht wurde von meinem Android-Gerät mit K-9 Mail gesendet.