idea: suggestion box

[Jeff Spaleta]

On Tue, May 11, 2010 at 3:35 AM, Paul Frields <stickster at gmail.com> wrote:
> If the submissions were called something other than "ideas" or
> "suggestions" we could avoid creating the impression that they'll just
> magically happen. "Proposal" seems a bit awkward but is probably
> closer to what people prefer across the project. It'd be helpful to
> include a simple checkbox along with each submission: "[ ] I am
> willing and able to spend a few hours a week working on this
> proposal." Tagging would provide a means to sort the proposals into
> clusters, possibly.


I already have a toy algorithm for similarity analysis. The Ubuntu
Brainstorm people gave me a dump of their anonymous voting record with
which to test a while back. I could only take it so far because its
anonymous. I did see clumps, but I couldn't actually identify a group
of people to talk to. And the way they construct polling doesn't
encourage people to build up a broad voting record so their data was
very sparse.  Instead of thousands and thousands of questions which
only get a few votes each on average. What I need is a system that
encourages people to express their interest consistently across many
tens of questions.  It doesn't have to be all questions, and it
doesn't have to be a simple yes/no..range works.   Enough people
answer enough of the same questions in a way that expresses like or
dislike.

The direct "i'll work on this tagging" isn't strictly needed. Whether
that sort of check box is there is not important to the broader
similarity analysis.  The clumps of people (and ideas) fall out of the
voting records. As long as many of our contributors..both potential
and active are building up a record of interest across the question
space...we get a dense enough data set to do a similarity analysis and
find the clumps fall out of that.

I would actually require people who used such a system to affirm a
desire to actively participate in some part of the project and to
state what they think they can contribute in terms of time on a weekly
basis on average working on "something".   The goal for me would be to
use the voting record to identify good fits between groups of people
and projects (both existing and new).  When on review a strong clump
is a set of new ideas and primarily new people, the Board may feel
compelled to step in and guide that group into standing up as a new
project team.

People who already have a strong idea of what they prefer to work on
are going to gravitate towards that on their own regardless.  The
similarity stuff asks to get a broad view of likes and dislikes to
help point people who are interested generally into the direction
likely to be a positive experience for them, and the others in the
team.

-jef