Hi,
On Fri, Dec 7, 2012 at 6:09 PM, Vincent Massol <vincent(a)massol.net> wrote:
On Dec 7, 2012, at 4:47 PM, Sergiu Dumitriu <sergiu(a)xwiki.org> wrote:
On 12/07/2012 07:56 AM, Vincent Massol wrote:
>
> On Dec 7, 2012, at 1:47 PM, "Ecaterina Moraru (Valica)" <
valicac(a)gmail.com> wrote:
>
>> Hi,
>>
>> The results for the #bugfixingday #8 held on 6 November 2012 are at
>>
http://www.xwiki.org/xwiki/bin/view/Blog/BugFixingDay+December+2012
>>
>> Great job everyone for closing 43 issues (and fixing 13).
>>
>> The next #bugfixingday is planned for 3rd of January 2013 (first
Thursday
>> of every month).
>
> Yes many thanks to those who participated.
>
> We still have 45 bugs to be back in the green though….
>
> I have the feeling that we should intensify a bit the BFD and I think
it would
be good to be doing this once every 2 weeks so that we're in the
green as a rule instead of the opposite being the norm…. WDYT?
I'd be a full +1 for this, except that I find that when something
happens too often, people tend to lose interest and the participation
will decrease. In the end, the marginal number of bugs will stay the
same, it's just the frequency that will change.
Yes I know. We need to find the good frequency so that devs don't loose
interest and participation is high enough. But ATM some devs could not
participate because they had something planned for the BFD day #8 and if we
have more maybe overall we'll have more participation?
I've racked my brain and I don't have a better idea for being in the green
than BFDs at the moment…
Well, another option is that we decide to systematically close bugs (not
new features/improvements/etc) that are older than say... 2-3 years. I have
noticed in this latest BFD that there are a lot of bugs that are no longer
reproducible and most of them are for the old versions that don`t even
behave the same way and that most of the stuff got rewritten in the process.
If we like, we could try to improve that and (maybe reduce the window to
1-2 years but) only close issues that have no comments/reactions to them.
Alternatively, there's the thing that Fedora does (and probably others)
where they systematically close bugs that are older than 2 versions
(basically older than the stable version that they support).
I guess there could be other ways that work for getting rid of old and
invalid bugs that are polluting jira.
Of course, this does not address active and valid bugs (since they need to
actually be fixed), but I think it would be a step towards the right
direction.