Collecting feature requests

[Luke Schierer]

On Sat, Jan 03, 2009 at 11:53:21PM +0200, Felipe Contreras wrote:
> On Sat, Jan 3, 2009 at 11:08 PM, Luke Schierer <luke at schierer.org> wrote:
> > On Sat, Jan 03, 2009 at 10:51:16PM +0200, Felipe Contreras wrote:
> >> On Sat, Jan 3, 2009 at 10:02 PM, Ka-Hing Cheung <khc at hxbc.us> wrote:
> >> > On Sat, 2009-01-03 at 13:14 -0500, Daniel Atallah wrote:
> >> >> On Sat, Jan 3, 2009 at 9:00 AM, Luke Schierer <lschiere at pidgin.im> wrote:
> >> >> > The trac plugin at http://www.trac-hacks.org/wiki/VotePlugin looks like
> >> >> > it would do this.  However, it requires an environment upgrade, which
> >> >> > seems to crash when I run it.  Perhaps Daniel can figure out what it is
> >> >> > that I am missing.
> >> >> >
> >> >>
> >> >> I got it working.
> >> >>
> >> >> You can now vote on tickets.
> >> >
> >> > I definitely prefer that to brainstorm. Would be even better if the
> >> > voting arrows are bigger and/or nearer the description.
> >>
> >> I prefer brainstorm because it provides better visibility, and it's
> >> easier to vote and add new items.
> >>
> >> It would probably model what users really want more effectively than
> >> votes on tickets, and proof of that is the input it already has
> >> received.
> >
> > The fact that it has a few hundred votes does not prove that it models
> > anything better than tickets could. It really hasn't told us
> > *anything* we wouldn't have known without it.  Just that QQ isn't all
> > that popular, and that otherwise file transfer and vv capabilities are
> > important to some users.
> 
> It tells you that people like voting systems.

Which is why we are implementing the voting plugin, and trying to see
what parts of brainstorm we can implement in Trac.

> 
> You can speculate about the validity of the results, specially in a
> short time span, but I bet the brainstorm would be much more popular
> than tickets, and that's the tendency you can already see right now.

Not really, no.  What I see here is what you mentioned just before, that
people like the idea of voting on features.

> 
> Statistically speaking, the bigger the sample, the more accurate the model.
> 
> > In fact, based on just *how* negative QQ is, I'd almost be tempted to
> > say that the site has been visited by a mad click happy person.
> 
> You mean a person creating multiple accounts just to vote up something
> that was specifically said not to affect the direction development
> happen? I find that unlikely.
> 
> -- 
> Felipe Contreras

I *highly* suspect that most users voting have disreguarded, never
assimilated the idea that their input does not affect the direction of
development.  

If brainstorm is in fact restricting users to one vote up or down per
item, then what is most clear from those results is that users don't
even really understand what they are voting on.  Because a vote down for
voice and video in QQ is hardly, if we were to follow the trends, would
hardly advance voice and video support in any other protocol.

Luke