Wikimedia Discovery/Meetings/Portal roadmap 2016-03-04

From MediaWiki.org
Jump to navigation Jump to search

Portal Team Roadmap Discussion[edit]

2016-03-04

Deb, Dan, Jan, Julien, Moiz, Kevin

Document under discussion:

https://meta.wikimedia.org/wiki/User:DTankersley_(WMF)/Sandbox/Roadmap

Framing from Dan:

  • This is a discussion, not a time for specific decisions
  • Roadmaps should be somewhat clear for the first quarter out, and less clear beyond that
  • Things can change if there are reasons; this is not a firm commitment

Deb

  • Be able to explain why we're making the changes we plan to make

Q3 Goals

  • These have already been discussed, so no surprises

Q4 Goals

  • Q4 goals have been cast as user stories
  • Clarification: The first Q4 test describe a test we're already planning to do in Q3 (there was confusion about what the devs implemented for the upcoming test)
  • ACTION: Combine this test with the one in Q3
  • ACTION: In roadmap, not goals, mention translating the tagline into more languages
  • Second Q4 test: Move the most important links "above the fold"
  • Third Q4 test: Streamline the page by combining all the language links into a language selector
  • Note that requirements for a test vs. production can vary widely (e.g. translations, op requirements, etc.)
  • ACTION: Change "4" to "3" in the last/production goal bullet
  • Fourth Q4 test: Stretch our imagination 
  • ACTION: Goals that rely on portallabs should explicitly mention that

Moiz issues:

  • I want the roadmap to be a bit more inspirational. 
  • Q4 seems like too many goals (5). Maybe we should have the same number of goals as this quarter, not more
  • These are all good goals, but that's my big concern
  • We could have "soft goals". Maybe 3 that we plan to do, plus 2 that we hope to do if we have time

DT: Note that goal 1 is being removed, so there is just 2 tests, portal labs, and push to production

KS: Lots happened in Q3 outside our control, so maybe our Q3 goals were realistic (even though we won't hit them)

JD: The tests in the portal labs goal rely on portal labs, so that's what we should focus on

DG: Need to clarify purpose of portal labs. It's not suitable for A/B testing. It is a place to obtain real user feedback on experimental ideas.

MS: Every quarter, there are external issues that affect us. We should expect them. 

DT: Do we want to scale back our goals? 

MS: I think we should pick 2-3 goals maximum

KS: Speaking generally, goals should be slightly aspirational; expected to hit maybe 80%, not 100%

MS: Team morale has suffered. We have been panicking at the end of the Q. We should have something to celebrate. If we have extra time, we can aim for internal soft goals after completing 

JG: Agree with Moiz. There is too much here. Difficult to estimate what is there. Maybe official goals and unofficial goals (more ambitious, less realistic, exciting)

DT: Either we describe all these great goals, but don't deliver all. Or we describe few goals, and over-deliver. What about having the A/B tests and production as the official goals, and the labs work as our stretch. 

JG: It's hard to get time to work on the exciting stuff because the basic stuff is time-consuming

MS: I would like the roadmap to be inspirational for the team, and to build positive (external) energy around the team. Have other teams read the goals and feel inspired. These goals are very specific, which is probably good for goals, but maybe this team needs something different?

DT: For Q4, would we rather work on tech debt such as localization or do we want to focus on labs page? Pushing last Q3 test into prodcution will require a lot of work. So the choice seems to be either A/B tests, or labs. 

JG: 3 options: Localization, tests, or labs. Localization is a large effort that would affect our ability to do the others.

DT: So localization would be the most bang for the buck?

JG: Depends what we want to do. If we have more A/B tests down the road, we should invest in the infrastructure earlier. I'm not sure what we want to do in the long term. Might take 2-6 weeks (JD: or more). It depends on what tests we want to run. Moving the search box would be quick and easy. But the language selector would probably take a lot of time

DT: Are #2-4 things that we do want to do, regardless of timing?

MS: Yes, but I would like to see them turned into goals a bit differently. When each gets done would be a decision for the team.

DT: Right. I want us to agree that these are the things we want to do. Next would be to agree on the order. 

JG: Yes these are things we want to do. But when? When ideas turn to goals, the question is are we comfortable achieving that in a quarter? 

DG: We're at time, and probably don't need to do answer this right now

MS: We should have an etherpad vote to figure out the priority/sequencing. Let's have another meeting. 

DG: Seems like the objective of getting to the point where Deb can present this as a tentative plan next week has been achieved. 

ACTION: Deb will schedule next meeting (probably Tues 9am)

ACTION: Figure out the voting process before that next meeting (Deb+Kevin)