Setting the sights

[Paul W. Frields]

The topic of our multiple task listings came up in the meeting.  No,
wait, that sounds like it just happened by itself; I raised the issue in
the context of our contributor deficiency.

In an only partially Swiftian moment, I suggested that we cull our task
list, wiping clean any tasks that we can't get done with the people and
resources we currently have.  After a little further thought, I propose
that we get rid of things on the list in the following order.  This is
just a proposal, and I'd be happy if people cared enough to argue about
it one way or another.  I'm prioritizing the list only so people can
easily respond to any level they find objectionable.

1. Any unstarted task targeted for FC-5 or before which has no owner.

2. Any unstarted task for which we have no contributor who knows how to
complete the task, regardless of ownership.

3. Any started task which has no current owner.

4. Any started task for which no updates have happened in >6 months,
regardless of ownership.

I know some people will find this very upsetting, and that's probably
good.  Maybe the idea sucks.  But on the other hand, maybe a list full
of tasks that aren't clearly going somewhere cause confusion and
consternation to new contributors.  If the task list we're working from
doesn't have jobs with a clear "howto," we're not likely to get anyone
to magically take those jobs over.  

The Docs Project is only going to thrive with the active participation
of people who may not have the technical mastery to take care of
multidisciplinary tasks.  Therefore, we need to reduce the number of
those tasks and skew the list more toward tasks that newcomers can do.
Any task that requires ownership and for which there is not already an
easy "howto" document for working on it, needs to have one.

We can move forward and encourage what I somewhat frivolously called a
culture of success in Sunday's meeting.  But to do that, we have to be
willing to pump out some bilgewater.  Why do we want to track a huge
number of tasks that no one is volunteering to do?  Is that really a
worthwhile pursuit, or is it creating unnecessary drag?

Speaking of which, I would also like to move for using Bugzilla for task
tracking, since (1) it's already there, (2) it has several components in
place in the "Fedora Docs" product we can use for any tasks that come
up, and (3) the rest of the project is using it.  (If the task tracker
changes for the rest of the project, any migration can include our tasks
as well.)  This may be a separate thread, though.  If anyone wants to
spin it off, please feel free.

I have done a minimum of second-guessing and re-editing myself above, so
that I get this out of my Drafts folder immediately.  If anything above
comes off as offensive or mean-spirited, I heartily apologize, since I
did not mean it to be so.

-- 
Paul W. Frields, RHCE                          http://paul.frields.org/
  gpg fingerprint: 3DA6 A0AC 6D58 FEC4 0233  5906 ACDB C937 BD11 3717
           Fedora Project: http://pfrields.fedorapeople.org/
  irc.freenode.net: stickster @ #fedora-docs, #fedora-devel, #fredlug