Jump to content

Wikimedia Discovery/Retrospective 2015-07-13

From mediawiki.org

Review of previous retrospective action items

[edit]

(NOTE: These were not reviewed during the 2015-07-13 retrospective itself, but I am documenting them here because in a perfect world, they would have been.)

Kevin

  • Schedule our next retrospective Yes Done
  • Clear blockers more quickly/frequently -- e.g. designate a scrummaster, or hold more frequent standups. Yes Done (hopefully).
    • Kevin is the designated blocker-clearer
  • Process for pulling tasks into sprints has not been communicated clearly
    • Minor progress, but far from resolved
  • Kanban (not Scrum) process has not been communicated clearly
    • Minor progress, but far from resolved
  • Everyone should make sure their contact info is up-to-date Yes Done (hopefully)
  • Didn't get David scheduled into the retrospective
    • It is not clear where this failure occurred. Will watch closely moving forward.
  • Stacking meetings on one day works well for some people, but not for others
    • No change, but meeting schedules are likely to be overhauled soon anyway.
  • Lack of IRC norms regarding buffers/bouncers, and whether people read scrollbacks
    • No progress.

Dan

  • Confusion due to having engineering work in data workboard Yes Done
    • (workboards are now people-centric, so engineering work stays in engineering workboards)
  • We are working on a lot of things slowly, rather than a few things quickly
    • Ongoing discussions, but no resolution.
  • Data Analysis is bushed; ongoing tasks suck up too much bandwidth to allow for long-term planning Hopefully better now
  • Protecting Oliver and really everyones time from randomization tasks
    • Hopefully better now

Wes/Tomasz

  • Guiseppe has represented us in ops, but we might need more Much improved.
    • Ops has committed fractions of 4 additional people to help us meet quarterly goals.
    • Ongoing discussions about getting more dedicated support
  • Some information droppage (notably when Wes/Tomasz are out simultaneously, or Dan by himself)
    • More of something to keep an eye on moving forward
  • Few of us sitting close to each other Yes Done
    • (SF folks now have desks near each other)
  • Budget has been a moving target
    • More of something to keep an eye on moving forward

Nik

  • Two search outages recently
    • Reports were written up
  • Nik is a single point of failure
    • Improved and ongoing. Nik has actively been transitioning knowledge to others

What's worked well:

[edit]
  • Quarterly review in general (this used to be, I kid you not, 20+ hours work for the product owner to put together a 60+ slide deck of every single thing done in the quarter, for a 3 hour presentation with half the organisation in the room. This time the expectations were clearer and it was much, much easier.)
  • Onboarding template was 80% of totally awesome (compared to 0% at most new jobs) (+1)
    • Folks have been updating the onboarding docs
  • Nik has been handling his departure well
    • Org's offboarding materials seem pretty good
  • Wikimania presentation runthrough generated a lot of really helpful feedback
  • Lots of improvement in process since the team was formed
  • Dan getting a headstart on Q2 planning. (he somehow feels more organized than he has ever felt!)
  • Terry, Lila, others feel like we're doing good work
  • Dashboard is generating really interesting conversations with nontechnical staff, e.g. Communications, Community Engagement
    • Frequently asked question is "Why is this data inconsistent across platforms?" and my response is "Because the platforms are inconsistent with each other" :-)
  • Data analysis pipeline, i.e. actually having a data analysis pipeline, continues to work really well
  • Strong focus on quarterly goals is a really good thing. Still room to improve in the future.

What could be improved:

[edit]
  • Prior presentation of quarterly review (show it to the team before presenting it to mgmt)
  • maps and wdqs: not moving fast enough to warrant weekly backlog grooming
  • Phabricator tickets arn't often very specific, leaving many questions when it comes time to implement. The tickets in the sprint board backlog should be immediatly actionable.
  • We don't seem to have a smooth way to convert Dan's "I want X" (a.k.a. user stories) into well-defined tasks + + +1000 <------ Ths one
  • Establish {sprint/iteration/weekly/whatever} goals in the form of user stories
    • Review success/failure to achieve them
  • Perhaps our process should shift to be more Scrum-like + +
  • Possibly tasks are larger than they could be
  • Still too few people in the office on team meeting days
    • When possible, try to get some face-time with the team
    • Moving whole-team meeting to Tuesday would help (or move others to Monday)
  • Ongoing work toward more clearly and concisely explain what we are doing
  • Could improve the terms we use internally (some goals seem ambiguous)
  • Maintenance/technical operational work (restarts, adding extra data centres) continues to crop up and detracts from achieving goals. Need clarity around this,.+
    • Unclear line between us and ops responsibilities
    • Should we designate a Discovery eng as our ops-focused person? (since we didn't get an embedded ops person due to budget reductions)+
    • Would be nice to spell out what we expect from them, and what they expect from us
    • Should it be more defined to be able to automate deployment?
      • Automation can be difficult when the underlying process keeps changing
    • Packaging and puppetization are forms of automation
    • I feel like we spend more staff time/money talking about dealing with Ops than we would if we just hired an Ops person. A wise man once said this to me. +1 <------ Ths one +1 +1
  • This retrospective (not as broad participation as we would hope)
  • We have too many short meetings...fewer longer meetings might be more productive

What confuses us:

[edit]
  • How can we interact more with the community?

Retro of retro

[edit]
  • Last 15 minutes was best
  • Circulate etherpad ahead of time
  • Nice balance of what went well and what could be improved
  • Monthly feels like a good cadence


Action items

[edit]

Voted up

[edit]
  • We don't seem to have a smooth way to convert Dan's "I want X" (a.k.a. user stories) into well-defined tasks <--Kevin will faciliate
  • Maintenance/technical operational work (restarts, adding extra data centres) continues to crop up and detracts from achieving goals. Need clarity around this
    • Wes will try to get us a dedicated Ops person. This is a long term plan.
    • Kevin will document Who from ops is helping with each of our projects
  • Experiment with Scrum <-- Kevin will get with James, Dan and other volunteers
    • This is likely to include changes to meeting schedules

Others

[edit]
  • Circulate retrospective etherpad ahead of time (Kevin)