Re: Enable voting on Bugzilla?

[Ryan McDougall <ryan mcdougall telusplanet net>]

On Sat, 2004-03-13 at 12:41 -0800, Rob Adams wrote:
> As long as I don't get an email with every vote, I don't really see any
> reason why it would be harmful to have the possibility of voting.  One
> piece of information that can be sometimes lacking is how many people
> are affected by a particular problem.  Serious problems are typically
> duped a large number of times, so these are easy to spot.  But sometimes
> I worry that conscientious bugzilla users who diligently search for an
> existing bug and find it already there have no way of reasonably
> indicating that this is a problem that affects them as well.  This may
> be especially problematic for possibly obscure issues like multihead,
> pseudocolor, layout switching, accessiblity, etc. that typically affect
> only a small subset of users, but are nonetheless important to fix.
> 
> On the other hand, as someone involved with metacity on a day-to-day
> basis, I am all-too-familiar with the fact that sometimes decisions have
> to be made that can be somewhat unpopular.  I'd hate to think how many
> votes we'd have on edge flipping or no-focus-on-map preferences.  For
> this reason, if such a system were to be implemented, I think that it
> would be useful if voting were disabled on resolved or closed bugs.
> 
> I would prefer that we didn't call it "voting" per se.  A "metoo
> counter" is perhaps more accurate.  But we do need to be quite explicit
> that simply voting me too on a bug does not ensure that any particular
> change will be made in the future.  With free software, the best way to
> "vote" is with a patch.
> 
> Even if for a particular maintainer the voting is entirely ignored, it
> gives people a widget to play with to feel like they've been noticed,
> like the button for crosswalks at most intersections these days.
> 
> -Rob
> 
> On Sat, 2004-03-13 at 13:33 -0700, Ryan McDougall wrote:
> 
> > On Sat, 2004-03-13 at 21:11 +0100, Fernando Herrera wrote:
> > > Sat, Mar 13, 2004 at 06:19:45PM +0000, Chris Howells escribió:
> > > 
> > > >Votes are used as a useful indicator for developers to judge what 
> > > >bugs/features are most annoying/wanted, and to prioritise their time towards 
> > > >those if they _want_.
> > > 
> > > 	What about the voting universe? Who is voting? Hackers? Advanced
> > > users? People reading /.? Home users?
> > > 
> > > 	If the population voting is not homogenous with the real user
> > > base results are not the real wishes of that user base, and they are
> > > not accurate.
> > > 
> > > Salu2
> > > 
> > 
> > The "universe" (ah set theory in a bugzilla discussion :)) of users are
> > those who have taken the time to register themselves with a valid email
> > on the bugzilla system, and thus generally represent the core of bug
> > reporting users. You'd know this if you stopped by the KDE bugzilla, or
> > read the link in the original post.
> > 
> > As someone familiar with mathematics, you should know that accuracy is a
> > relative measure. Political polls use very strict statistics to arrive
> > at very high accuracies given they only asked 0.01 percent of the voting
> > population.
> > 
> > Cheers,
> > Ryan

Rob basically made they exact same point I want to make: so long as its
not hurting anyone, I think it should be enabled. I don't want to
degrade the signal-noise ratio, but I am as yet unconvinced that a "me
too counter" would necessarily do so, so long as its clear to users that
"voting" is not binding. As such, I think this is an important
discussion to have.

ps. Including gnome-bugsquad

Cheers,
Ryan