Hi everyone,
I had intended to send this email prior to our weekly meeting this
week, but I ran out of time to send it. I believe I covered all of
the major points in our meeting, but I'll send what I had drafted
anyway just in case I missed something important.
As you all know, we've had a bit of a muddled planning process leading
up to our quarterly reviews, which has evolved to be something like
the following:
1. We have an ideas list page[1]
2. About a month out from the quarterly review, I ask everyone to put
what they would like to be working on in a spreadsheet broken down by
month[2]
3. We have discussions about what the "big" project should be
4. We refine that into a wiki page[3], consolidating and pruning the
spreadsheet
5. We boil that down into a slide deck, and present our plan at the
quarterly review.
One big problem with that process is the ideas list is a vague,
undifferentiated list of random items, some hugely important, some
trivial, and some too vague to assign priority to. Another big
problem is that it's unclear to others exactly how to influence our
plan. It's probably even unclear within the team.
Dan, Howie and I met earlier, and we'd like to propose something a bit
different. We'd like to achieve some important improvements:
* Have a well-groomed, well-vetted backlog of possible projects with
basic priority levels (e.g. high, medium, low) and reasonable
introductory descriptions for the high-level projects.
* Hold ourselves accountable to getting the non-emergency things we
work on on the list
* More methodical gathering of input from other parts of the organization
How we hope to accomplish this is to have a regular (probably
fortnightly) meeting to flesh out high priority ideas and triage low
priority ones. We'll make sure everything that is listed as high
priority meets a standard of completeness with respect to its
description and rationale. We'll more explicitly solicit input from
other parts of the organization on what the most important things are
from their perspective. We'll then consider only those things the team
agrees are high priority as work for the coming quarter, and use our
existing process to narrow things down to the final plan.
Dan has posted the proposed process to
mediawiki.org, which you can read here:
https://www.mediawiki.org/wiki/Wikimedia_MediaWiki_Core_Team/Backlog/Proces…
The actual backlog would live here:
https://www.mediawiki.org/wiki/Wikimedia_MediaWiki_Core_Team/Backlog
Dan will lead this process, but the actual decision-making will be
team consensus driven by Tim and Dan. If there's a hopeless
stalemate, I'll step in with an arbitrary, ill-informed decision that
everyone will hate. Don't make me do it. :-)
Let's do some on-wiki tweaking/discussion during the week, and then
finalize something at our weekly meeting next week.
Make sense?
Rob
[1]
https://www.mediawiki.org/wiki/Wikimedia_MediaWiki_Core_Team/Ideas_list
[2] MW Core Allocations sheet:
https://docs.google.com/a/wikimedia.org/spreadsheet/ccc?key=0Agte_lJNpi-OdE…
[3]
https://www.mediawiki.org/wiki/Wikimedia_MediaWiki_Core_Team/Quarterly_revi…