Community V. "Community

[Colin Watson]

On Sat, Mar 21, 2009 at 03:22:25AM +0100, Mario Vukelic wrote:
> On Fri, 2009-03-20 at 10:14 -0300, Derek Broughton wrote:
> > it's not that they're shutting us out, it's that they don't even seem
> > aware that they're missing a very valuable source of input to their
> > decision-making process.
> 
> I can very much agree with that, but I have the impression it's very
> much a people power problem.
> 
> And here is an opportunity for "mere" users to help with that and get
> appreciation: read the ubuntu-users archives and summarize the main
> issue users express there, then transform them into pieces of
> information that developers can work on (blueprints, wishlist bugs, what
> have you).

(Blueprints are software design documents and should be written by
software designers in order to be effective; but users can certainly
collate popular ideas and pass them on as something developers can run
with.)

> I just think it's unreasonable to expect devs to do that.

It's not entirely unreasonable, and we ought to do better at it; but we
do need to be a bit careful where we set the bar. User lists inevitably
involve a lot of duplication, simply because users can't be expected to
follow the history of everything to realise that their question has been
asked 20 times before. I would rather that Ubuntu developers were, by
and large, spending their time on writing code and fixing bugs in Ubuntu
rather than on reading mailing lists (for example I generally only read
this one while waiting for time-consuming tests to run, as now); but
perhaps if somebody were doing an effective job at drawing our attention
to major issues currently occupying the collective mind of the users
list, it might be easier for developers to respond efficiently without
drawing too much time away from the work that only they can do.

IME posts or bug reports that say "hey, a dozen people over on the users
list or on the forums are having trouble with something and I think it's
an important bug, please can somebody have a look at it" are a good
indicator of signal and tend to get looked at ahead of problems that may
be isolated. After all, it's in our own best interests to work on bugs
that affect lots of people, since that way we tend to get fewer
duplicate bugs!

The test should not be that posts on the users list routinely get
developer responses; the test should be that, in the aggregate, problems
raised on the users list are passed to developers and addressed (whether
that be by fixing bugs, adding documentation, explaining why it's
impossible, or whatever). The more efficiently this information flow can
be handled, the better the chances that problems will actually be
addressed.

Brainstorm is an input into the decision-making processes for
Canonical's Ubuntu development team; obviously it is not the only one
but it does get attention. We're still getting used to doing this
efficiently, and obviously not everything can always happen in strict
order of greatest number of votes. ubuntu-users could be an input too if
we figure out a way to do so reasonably efficiently, perhaps by making
sure that ideas raised there are represented in Brainstorm or as
well-described bug reports so that we have a way to weed out cross-forum
duplication.

-- 
Colin Watson                                       [cjwatson at ubuntu.com]