Use of CLOSED in Bugzilla (was) Re: [Bug 284853] ...

[Aleix Pol]

On Mon, Aug 27, 2012 at 11:55 PM, Adrián Chaves Fernández
<adriyetichaves at gmail.com> wrote:
> I think in projects with no QA team, the reporter can be the QA:
>
> - Reporter opens.
>
> - Developer resolves.
>
> - Reporter confirms (closes).
>
>
>
> O Luns, 27 de Agosto de 2012 23:35:20 Martin Klapetek escribiu:
>
> On Mon, Aug 27, 2012 at 11:26 PM, David Edmundson
> <david at davidedmundson.co.uk> wrote:
>
> Can we continue this discussion off-bugzilla and onto our mailing
> list. It's quite interesting, but not really related to that
> bug-report.
>
> CC-ing the KDE-Telepathy (where this started), and KDE-Quality mailing
> list (where this discussion should be)
>
> Backstory for those new:
>  - there was a comment "is this bug closed" on a bug that was marked
> as resolved. In KTp we're currently treating those two states as the
> same thing.
>
>>Well, I meant CLOSED as opposed to RESOLVED. I guess you are not using
>> CLOSED
> (yet?) either.
>
> As I understand it in bugzilla terms, resolved is the development team
> saying it's fixed. Closed is used when a QA team has accepted the
> patch.
>
> Is this correct?
>
> Given manpower/resources, I was under the (wrong?)  impression that in
> KDE we simply go to resolved and closed was in reality
> unused/pointless. Certainly with my developer hat on, it seems it.
>
> 1) How _should_ we be using "closed" as opposed to "resolved" in an ideal
> world?
>
>
> Also note that bugzilla doesn't allow setting CLOSED directly, it needs to
> be RESOLVED and only then can it be CLOSED. Which for our project (KDE
> Telepathy), which has no real QA (assuming the usage example above holds),
> causes unnecessary overhead, needing to "close" the bug twice, which in turn
> leads to creating (useless) email noise. In other words - I see no benefit
> of CLOSED in small projects like ours.
>
>
> --
>
> Martin Klapetek | KDE Developer
>
>
>
>
>
> _______________________________________________
> Kde-testing mailing list
> Kde-testing at kde.org
> https://mail.kde.org/mailman/listinfo/kde-testing
>

I don't think we should expect the users to actually report back to
the bug after being fixed.

With the current state, distributions don't let the user make sure the
bug is fixed until months after the bug report was closed. I don't
think anybody wants to go back to those things...

If a project wants to diferenciate between FIXED and CLOSED, then it's
fine, but I think that FIXED means that the bug is solved until proven
wrong. (note that nobody can know if the problem is ever completely
gone... unless the feature has been removed).

Aleix