Notes from discussion

Agenda:

  • Intro to workshop (5 mins)
  • Discuss the findings we are working with (30 mins)
  • Intro to activity (10 min)
  • Concept organization and generation with break out teams (60 minutes)
  • 15 min break
  • Breakout teams present (25 mins per team)
  • 15 minute break
  • Group discussion (15-30 mins mins)
  • Cluster ideas into groups in a new mural (10 minutes)

Links to things within the initial slides:

  • Wikipedia:Guide/Copyright in practice - https://cs.wikipedia.org/wiki/Wikipedie:Pr%C5%AFvodce/Autorsk%C3%A1_pr%C3%A1va_v_praxi
  • login with email - https://phabricator.wikimedia.org/T30085
  • who edited an article? - wikiblame - https://phabricator.wikimedia.org/T2639

Shared understanding of findings

  • JM: “Reach out to editors who stopped editing in this focus”? More about recruiting than retention. One of the things different on Wikipedia from Twitter, Instagram—complex workflows, lot of things you need to learn. Need to set the table before you send out the invitations. Once we have a better experience, focus on recruitment.
    • +1 Nick, Benoit, Alex, Joe
  • BE: Improve existing things. Vocab question: when we talk about “Wikipedia” is that an intentional limitation?
    • NQ: yes, we’re focused on medium-sized Wikipedias.
  • AS: For “expand tooling for Article Rescue Squadron”, I think one of the big goals of this project is to support and encourage practices where experienced editors support good-faith new editors.
    • JM +1 to supporting those who support ideas
    • NQ: It's a goal to support editors in these ways, but is it within the focus' we picked?
    • AS: Partially the conceptual part (#8).  Also #7 mostly and #5.
    • JM: #9?
    • AR: As we're defining what we do to support the experience of new editors, we'll see where we need to support-the-supporters. We should collect the names [?] of the mentors and supporters, so we know where to find them when we need to ask for their input.
  • RH: Profiles for audience builders is addressing #9. If they want to create, instead of telling them their contribs are useless, redirect them to their userpages. That might give them space/time to learn, and thus eventually convert them into productive contributors.
    • +1 AR
  • PG: Some ideas can be stronger or less strong depending on the framing. We should be as explicit as possible what the impact we’re hoping for is. E.g. reaching out to editors who stopped editing: depends on why we’re reaching out how well it fits under our findings (is it to make sure they understand the policies or know pathways for progressive growth?)

Google hangouts for breakout teams: Team A Team B Team C Team A presentation:

  • JM: We identified key issues rather than specific ideas:
    • Training
    • Subject-area grouping (like Wikiprojects)—could provide more constructive collaboration, more social links,
    • Communication and collaboration (live collaborative editing)
    • Giving tools for users to organize and recruit other users (particularly activists, who have cohorts and networks around their topic)
  • Discussion:
    • AR: Any ideas that work together?
      • JM: Definitely. If you can guide people into a subject-area group, the group benefits a lot.
      • NQ: More general ideas (“strengthening subject-specific collaboration”, “better communication”) are likely to be more useful than specific
    • RH: Did you focus more on audience builder or social changer? Are some personas going to be the focus more than other?
      • NQ: Social changer. Yes, some personas (e.g. audience builders) are less important than others—structure of concept evaluation isn’t determined, but I think it will implicitly or explicitly reflect that fact.
    • JM: Our system is so complex compared to others (e.g. Facebook, Twitter). That’s one of the reasons I’ve become focused on people’s ability to communicate. I’m not sure we’ll ever have a help system—in story after story, success comes from another human intervening to help out. Want to make that happen more often.

Team B presentation:

  • PG: Started with the knowledge sharer. First step is about discovering the ability to edit. Ideas about clear messaging for newbies, asking the user for their intent before contributing, surfacing the number of contributors as an invite. Next step is about fixing outdated information years later: idea about suggesting edits based on real life activities. Next step is about editing anonymously. Ideas about invitations to create account, guided onboarding. Next step: tagged article. Ideas about self assessment, signed templates. Last steps: discovering policies, continuing to edit. Ideas about mentorship program, training by reviewing example edits, micromoderation.
  • PG: Next, reactive corrector. Ideas are a bit repetitive from the knowledge sharer. Site messaging aimed at newbies, asking users what their interest is when registering, surfacing impact on initial successful edits, getting experienced editors to send self-written note to new editors (reactivate the user during a period of inactivity), adjust welcome messaging to sound less robotic, notifying about new ways to contribute, getting people to mark possible mistakes and asking reactive correctors to fix them.
  • Discussion:
    • PG: tried to focus tightly on our user journeys.
    • JM: Do people have an idea of which phase is most promising to intervene at?
      • PG: based on clustering, I think that the big areas we identified were (1) letting people know they can edit (not relying on serendipitous discovery), (2) task recommendation and interest areas and how to participate in them, and (3) how to recover from problems, how to prevent discouragement
      • JM: +1
    • NQ: what ideas did you have around that 3rd area?
      • PG: signing templates (making clear they’re from a person—opening a channel to communicate with them), mentoring and training, having a working area where you can try to address problems, in-context editing comments and dialog. I’m sure there could be others as well.
    • AR: really like how there are a few concepts around both how to help people when they’re aren’t problems and how to help when there are.
    • BE: in all the concepts, there seems to be a fundamental assumption that you have experienced editors who are willing to be kind to be newcomers. People do bite newbies.
      • PG: any idea that involves human-to-human communication has the potential for problems. New users not getting responses from experienced users, experienced users not getting responses from new users. Not easy.
      • NQ: Some experienced editors are fundamentally bitey, but I think the majority are bitey because of circumstances (workload, no response when they try to be nice), which we can help shape.
      • AH: Take a behavioral economics perspective on biting: biting is an externality—how do we make it so that biting newcomers feels expensive to the new users who do the biting? Some obvious things: highlighting the kind of interactions people have (identify biters), identify and reward people who have good interactions.
        • +1 AR, BE
      • AR: We're working to solve NEE, but clearly also need to support the people who support new editors.
      • JM: How would we do that? Perhaps, like StackOverflow - the community votes up/down the answers given to questions. We can hope people will vote up the constructive/friendly responses. I.e. social pressure to be helpful.
        • +1 BE, AH

Team C presentation

  • AR: We had box checker and joiner-inner. Box checker chooses to edit Wikipedia because she’s a reader and feels confident, but doesn’t realize how unfamiliar you are. Ideas: help her discover visual editor earlier in her experience, make help more discoverable, make help easy to find and structured, provide choices of support routes, help her understand policies, provide insipration to come back.
  • AR: Joiner-inner. May be the least likely to evolve in a productive aspective. Ideas: structured profiles, edit recommendations, first contact at account creation (when you create an account, you are welcomed, usually by the website itself—here the idea is being personally welcomed by a new editor). She has a lot of support because she came in through a course—but might be opportunity to give more editing suggestions and help more discoverable. Surface the impact of her initial successful edits.
  • BE: One of the ways for people to feel more comfortable: suggesting edits.
  • Discussion
    • NQ: There are a lot of opportunities for people to contribute in social ways, so I think they do have long-term potential. What do people on the team think?
    • RH: The box-checkers and the people who assign them with tasks, could be provided with more material to teach others. Once they're done with the edits they have, they can move along and become mentors themselves.
    • AR: Teachers/mentors need better tools to educate their students on how to edit Wikipedia.
      • +1 BE
    • AS: People who want to be social—we need them in the movement. The strategy has a lot about looking beyond the edit count. If people start editing and get pulled into more social parts of the movement, that’s good. We need better ways to encourage and track that.
      • +1 BE
    • AA: Joiner-inner, who doesn’t edit much but likes participating in discussions. One potential role for people: discussion moderation. Not a role that really exists in any proper way. Very unstructured mediation procedures—not the same as discussion moderation (people who can spot toxic comments). It often doesn't take much to redirect small toxic comments. We could direct joiner-inners to that.
      • AR: Really good idea. Hadn’t thought of that. What kind of skill sets would that person need?
        • +1: BE
      • AA: People need to be somewhat bold and confident in telling others to stop. Might need admin tools. Might need structured discussions.
        • +1 RH
      • NW: Community health initiative working on training admins to deal with toxic situations.
    • NW: Dangers with box-checkers: can turn into “rule-lawyers”. Don’t want overenthusiastic new, inexperienced people who try to apply the rules overly literally.
      • NQ: Fits joiner-inner better?
      • NW: No, I was thinking box checker, maybe because of the persona name.
    • RH: Want to reiterate the point about providing teaching type material for people assigning tasks to box-checkers. It blends into the stuff Team A was talking about, better onboarding material. One teacher from the interviews was working on creating a wiki syllabus.
    • JM: Do you think the WikiProjects would appeal to the joiner-inners?
      • NW: It would slightly, but as far as I know all projects discourage off-topic discussion—they only want people discussing article content, not discussing the topic generally. Generally discourage, sometimes quite vigorously.
        • +1 BE (there’s the idea that “Wikipedia isn’t Facebook”)
      • PG: Related to whether topic-related focus could be useful beyond the online context, I think it has potential. E.g. events that chapters organize, those are sometimes around topics, e.g. women in science. These sometimes result in attracting people who are outside the usual newcomers.
      • AS: There are already educational handouts avail for teachers to use. Local affiliates and Educators are work on these. Connecting the content-face with the outreach-activities has been the problem. How does the infrastructure relate? [?]  We tried making this on the Library’s “Research Help” Portal, with specific boxes calling out areas of interest to particular demographics. https://en.wikipedia.org/wiki/Wikipedia:Research_help  (impact/oucomes: https://en.wikipedia.org/wiki/Wikipedia:Research_help/Pilot_report)
        • The resources linked above are being updated.
        • Outcomes from that portal:

General discussion

  • MC: Joe is working on a “facilitating subject interest”, could be combined with my “tailored tasks” groups?