Re: minutes of ESC call ...

Hi Tommy,

        Thanks for your mail ! :-)

On Fri, 2012-03-23 at 06:29 +0100, Tommy wrote:
> >     + about 1/2 of the calc regressions are already fixed in 3.5
> >             + 3.4.x will get no more updates
> >                     + no need to explain more, close them.
> > AA:         + review and re-close these / discuss in QA call (Rainer)
>
> I wonder what will happen to the still open 3.4.x most annoying bugs (23  
> bugs yet). https://bugs.freedesktop.org/show_bug.cgi?id=35673

        Good question - I guess one for the QA call later today (it'd be great
to have you there). I suppose we should (as you say) check that they are
all either fixed in 3.5.x and/or make them 'most annoying' for 3.5. I
guess volunteers to do that work (if that's what QA decides) are also
appreciated :-)

> > * 3.5 most annoying bugs ...
> >     + 58 (of 168): older 57/157 58/153 52/145 49/140 44/132 41/124 32/104  
> > 28/93 21/76 23/71
> >          35%              36%    37%    36%    35%    33%    33%    30%     
> > 30%   28%   32%
>
> I think we should look at the fixed/unfixed ratio in another way...

        :-)

> the percentage looks almost the same week by week just because new bugs  
> have been constantly added... also the number of unfixed bugs is rising
> for the same reason.

        Sure - and to some degree perhaps some of our most annoying bugs are
getting a little less annoying over time as people's tolerance for bugs
decreases (good) and they file and prioritise more.

> so these numbers could be misunderstood as sign of immobility or worsening  
> of the develepment cicle, shici is not IMHO.

        Right, IMHO we're fixing ever more of the most annoying issues that
have been found some of them really long standing nasties.

> what I consider most important and encouraging for the devs and users is  
> that the number of fixed bugs is growing, if you count backwards you have
> 110, 100, 95, 93  .... 48 fixed bugs.

        Yes ! that's a reasonable positive measure of progress.

> so the most important thing is that the fixed bugs numbers is rising  
> steadily and this should be put in first place!!!

        Certainly; on the other hand, my interest in publishing the numbers is
to encourage people on this list to consider working at fixing
regression and/or most annoying bugs :-)

        As such a rather negative way of looking at the numbers could help
motivate people ? ;-)

        One of our big needs here is for people to dig into the UNCONFIRMED
bugs - that havn't had an independent confirmation from people - such
that we can (more quickly) dig out regressions and new issues from the
large incoming set of bugs. A quick bugzilla query for libreoffice bugs
in the unconfirmed state gives a good chunk of useful busy-work to fill
even small gaps of time :-)

        Thanks,

                Michael.

-- 
michael.meeks@suse.com  <><, Pseudo Engineer, itinerant idiot