What we have learned so far on ubuntu-release upgrader hugday

[melchiaros]

Am 22.03.2013 16:08, schrieb Brian Murray:
> On Fri, Mar 22, 2013 at 12:55:00PM +0100, Alberto Salvia Novella wrote:
>> I thought this myself, since I found that most of the bugs I have triaged
>> myself were caused by this. On the other hand, I think other things we can
>> possible learn are:
>>
>>
>>      - Confirmed bugs are less likely to be triaged, since are intended to
>>      Ubuntu BugControl members. Perhaps automatically allowing some kinds off
>>      users to access triaging bug controls will be a good idea, also because
>>      most bugs are never assigned priority.
>>
>>      - Although I think the bug day is a good idea, we were only four users
>>      participating on it. This warns me the Ubuntu community isn't so active. I
>>      also study computer engineering, and I saw many people stopping using
>>      Ubuntu in favour of Windows since they perceived Ubuntu as unstable and not
>>      compromised with freedom. Improving bug handling an incursion of new users
>>      in it seems to be a high priority, perhaps also giving the users the
>>      opportunity of not sowing non free applications in the software center and
>>      deeply explaining why Canonical chose to include them in the system.
>>
>>      - Many users weren't using the ubuntu-bug command to report bugs. Perhaps
>>      it should be clearly stated in some visible place that manually reporting
>>      bugs in Launchpad is for when is impossible to use this command.
> This is done by having the "Report a Bug" pages in Launchpad redirect to
> the wiki page at https://help.ubuntu.com/community/ReportingBugs.
> Additionally, in most cases the ubuntu-release-upgrader should help you
> file a bug report (using apport) or by telling you to use 'ubuntu-bug
> ubuntu-release-upgrader'.
>
>>      - Editing the wiki took many of time used in the Bug Day, so perhaps
>>      creating good automated tools for it will increase noticeably the quantity
>>      of bugs that can be handled.
> I'm looking into the hugday tool and from what I can tell so far there
> is an authentication issue, for me at least, when trying to close bug
> reports.  I'd welcome any help with this though.
>
>>      - In the incomplete bugs section, some of them needed to be converted to
>>      question. I think this action should be contemplated as valid to mark the
>>      bug off, perhaps also in other sections.
> Yes, if the bug report is clearly not a bug report and more of a support
> question then converting the bug to a question is a valid action.  Do
> you have some examples of those?
>
>>      - The progress section shows nothing, so perhaps it can be removed.
> I believe that issue is specific to the ubuntu-release-upgrader package
> and is my fault, sorry about that!
>
> --
> Brian Murray
> Ubuntu Bug Master
>
Sorry Alberto, my Spam-Filter is a bit overmotivated. I have just seen 
your mesasage in parts through Brians mail.

Your point on the gab between confirmed and triaged is quite correct. In 
principle everybody on launchpad can do what BugSquad people do with or 
without membership.

I would transform your proposal a bit. In moment the BugControl has a 
hugh access barrier. The people are forced to present them to 
someone(well properly also Brian) and ask for permission. This is o.K.  
But also the other way arround may be accessible. For sure there are 
people outside that fullfill(also a properly) the requirements for 
joining BugControl. You see them again and again and over long time. It 
is also a way to just invite them for a less formal join.  The 
difference is the invitation, and who is not lucky to be invited.

Your second point Alberto is going into the wide. Ubuntu v.s. Windows 
and Canonical monetary approaches.... This is to much for what I have 
whant here. But one point is absolutely correct. On this one single time 
event hughday, the activity is very low. Only four people is much to 
less. I just whant keep this as observation. The hughday-tools may be 
one cause. Also may be, we should have given a special workflow for 
handling such bugs(like apt-clone usage).

To the third point is to say.  It seems to be especially the long year 
linuxer that try to report without apport. I have some time looked into 
the launchpad profiles, when I hit such a report.

May be the explanation is that they have done so in past and just are 
willing to do in futhur:)

I think a redirect is quite o.K. It is not o.K to force someone to much, 
even if that makes it easier to us. The consequences for them a much 
more impressive. A missing raring tag will not show their report for the 
actual development. Or when we see this they have also extra work. Or .. 
you know.

To resume:

I think we may have now tree points:

-the ubuntu-release-upgrader behaviour on xorg
-the gab between confirmed and triaged
-the non working hughday-tools

@Brian

Fabio has also written to look at the tools in one of the hugday mails 
on the list(AFAIR). So there is the first on helping out.