Bug#830344: How should the TC help with a project roadmap?

[Sam Hartman <hartmans@debian.org>]

Here are my thoughts on the road map and TC involvement.

There is value in two levels of thing:

* Goals that we've committed totrying  as a community.  For these, RC bugs or
  NMUing a package are valuable.

At this level it's desirable to have review of the plan to achieve a
goal.  It's frustring to make a bunch of stuff RC buggy only to later
realize that even if you had fixed those bugs you never would have
gotten to the goal.

I think that review needs to be positive--some people need to say yes,
not simply no one raises objections.  Also, it needs to be reviewed to
make sure all the stakeholders are involved.

* Second level: wishlist.  Things people think would be valuable.  Easy
  to add to.  May not represent project-level commitment to try for the
  goal.  You may not want people NMUing at this level.  As an example,
  removing build information from a binary package does sometimes make
  debugging harder.  If we're not actually going to achieve reproducible
  builds, it's not clear that making those changes is valuable.  (In the
  specific case of reproducible builds, we've met the bar already, but
  the point stands as a generalization)

----------------------------------------

TC skills that may help here:

1) Across the entire TC we have a moderately good coverage of things in
the project.  There are probably gaps.  Across the TC we have fairly
good coverage of who to go to for more depth about a given issue.

2) We're builting a TC that's good at working with people and helping
facilitate communication.

3) We can do technical review for completeness of a proposal.

----------------------------------------

One area where I'd like to the see the TC help is to try and avoid late
stakeholders appearing.  That is, you put together a plan, start working
on it, and then discover late in the process that  you missed some key
player, and they disagree with your goal.  So you invest a bunch of time
and run up against a stone wall.

I think if we worked on it we could be fairly good at making sure people
have talked to a lot of the stakeholders they need.
I really hope the process supports that sort of review because I believe
it could significantly help with burnout avoidance.