Community Capacity Development/Community Research



Interviews with sixteen communities or countries[1] (predominantly in the Global South) surfaced seven community capacities for development, based on the strengths, challenges or needs reported by those communities. This report aims to define these seven capacities - to create a common vocabulary for future discussions - as well as to provide a non-exhaustive list of example actions for building these capacities:

  1. New contributor engagement and growth
  2. Community governance:
    1. Roles
    2. Policy creation and revision
    3. Conflict management
  3. On-wiki technical skills
  4. Communications
  5. Partnerships

This report is part of a phased approach to capacity development, where this report focuses on identifying and documenting key areas for development; it does not focus on building out a comprehensive list of solutions or next steps. Instead, the next phase will be to create separate, on-wiki spaces for each capacity, in which community members and WMF may discuss and clarify central problems, existing solutions (from other communities), and potential future work (e.g. research, trainings, facilitated meetings, etc.). As such, this framework will serve as the basis for deeper discussions across communities, and with specific communities. Moreover, while these seven capacities were identified after interviews with certain communities, the capacities and their future associated resources may be relevant across many wiki communities.

We invite your feedback! Please add your comments/questions to the talk page. If you’d like to engage further or keep updated, please add your name on the framework main page.

Background

edit

While community capacity development has been a long standing goal of the Wikimedia Foundation (WMF), there has been limited and at most sporadic work to understand, identify and define key capacity needs across communities. There have been a few initiatives that have addressed key areas - e.g. Evaluation, Organizational Effectiveness - but the breadth of capacity needs, as well as the specific needs of each particular community, remain relatively undocumented.

In order to address this gap, the Community Engagement department at WMF engaged in a series of community interviews,[1] to understand the following:

  • Environmental context. Key environmental factors that have helped or inhibited the establishment and growth of a wiki community. Many of these factors are outside of an individual's control, but will affect their ability to contribute either on- or off-wiki.
  • Community structure and norms. For example, governance (How are admins selected? Is there a de-sysopping process?), conflict management, policies (How is policy made or changed? Is current policy serving the project well?), availability of technical skills, on-wiki spaces, etc.
  • Programmatic capacities. For example, planned activities (e.g. events, programs, partnerships) and their associated operational activities (e.g. planning, evaluation, volunteer engagement), media relations, etc.

The research phase of this work had three primary goals:

  1. To understand the community’s strengths, challenges and needs, as identified by community members.
  2. After understanding key areas across and within communities:
    1. Aggregate common themes across communities, and specific needs within communities
    2. Identify common/scalable resources and solutions, and then collaborate with specific and interested communities to discuss possible community capacity building through WMF support (incl. funding capacity-building activities or services), mentorship or partnership.

Structure of the report

edit

From community interviews, eleven capacities surfaced as key areas for capacity development, but only seven are included in this phase of the report. The four capacities that were not included were: Planned activities, Volunteer engagement for planned activities, Learning & Sharing (including Evaluation), and Organizational Effectiveness.

Each capacity reported will have the following structure:

  • A definition of the capacity.
  • A summary of findings from the community interviews conducted so far. This report will only report findings at an aggregate level, with very few references to specific communities (and only with explicit interviewee permission).
  • An list of example actions to address this capacity. This list is intended to give concrete examples of the ways WMF might support, mentor or partner with communities, but it is not intended to be a comprehensive list of next steps.

Environmental context

edit

There were a number of environmental factors that community members identified which have helped or inhibited the establishment or growth of their wiki community.

While many of these are outside the Wikimedia mission or our ability to directly affect, some are addressable via partnerships with appropriate organizations (addressed in the Partnerships section). Below is a list of environmental factors that were identified by more than one community:

  1. Geographical dispersion of potential and actual contributors: Many emerging communities have significant global diasporas, while others operate across very large physical distances. As such, off-wiki meetups can be rare or non-existent.
  2. Language status and basic literacy: Emerging communities often have more than one widespread language, and as such, an individual’s native language and their language of education (or even language of literacy) may not be the same. With different languages used in different contexts, as well as varying levels of literacy in each language, “language status” and literacy have wide reaching of effects on an individual’s ability to contribute. The role and interest of academic / governmental institutions on native language proficiency varied between communities, with some communities reporting minimal or decreasing interest by these institutions.
  3. Information literacy and access to source material
    • Secondary sources are scarcer or less accessible to contributors in emerging communities. Furthermore, secondary sources simply do not exist in certain languages or topics, due to a wider range of historical reasons (e.g. language status, cultures of oral histories, etc.). Even daily newspapers and magazines (frequent sources for Wikipedia) are not always available in certain languages, or considered reliable by the community.
    • Information literacy (e.g. recognizing the difference between primary, secondary, and tertiary sources; judging reliability of online sources) is not generally taught by education systems, and so academic norms (including notions of plagiarism and copyright) are not as inculcated in some communities.
  4. Internet access and infrastructure: While internet access has increased dramatically, mobile access has emerged as the primary form of internet access in some communities (e.g. in Asia or Africa), with large sectors having low/no access at all (due to infrastructure or prohibitive cost). While certain communities / countries did report access to home or personal computing, computer access was primarily reported at work or at school/campus.
  5. Cultural norms: Many cultural factors affect contribution to the Wikimedia projects, both positively and negatively. Some examples mentioned were:
    • The relative lack of hierarchy among contributors is appealing in some cultures, but suspicious or uncomfortable in others.
    • The impersonal tone of many Wikipedia policy pages and templates may be perceived as challenging, "cold", or off-putting in some cultural contexts.
    • Notion of volunteerism and altruistic effort vary, or may not be well understood in regards to the Wikimedia projects.
    • Notions of knowledge and authority vary, affecting whether an individual feels empowered to contribute.
  6. Governmental / political climate: Governmental points-of-view span a wide range, creating both obstacles and opportunities. Some obstacles identified were outright suspicion (e.g. suspicion of civil society or self-organization outside state or religious groups), inhibiting governmental policy (e.g. on neutral point of view or freedom of expression on the wiki), and active intervention (e.g. promotion of particular points of view on local-language Wikipedias). Conversely, some governments have perceived Wikimedia projects as vehicles for education and language preservation, and have extended support to local volunteers (e.g. by providing language dictionaries or encyclopedias).

New contributor engagement and growth

edit

New contributor engagement and growth encompasses all activities that encourage new contributors to integrate with wiki-culture, persist and grow. However, it does not include outreach activities that encourage potential new contributors to contribute.

Summary of Research

edit

Engagement and growth activities reported fell primarily into four buckets:

  • Personalized welcomes: While bot-driven welcome messages were common, they were not ubiquitous, with some communities reporting the policy of not using bots to welcome new contributors. Multiple communities identified human communication as a great way to greet and welcome new contributors, with these communications happening either on-wiki - with personal welcomes from more experienced contributors - or off-wiki (but still online) through social media groups. Interviewees reported increased appreciation for these welcomes, remembering that first human welcome after their initial contributions.
  • Recognition: Peer recognition was a clear, consistent, and cross-community theme, with a range of implementations. Barnstars, wikilove, the [thank] button, physical tokens of appreciation, and even encouragement to hold sysop positions were all reported as effective ways to encourage and retain new contributors.
  • Mentorship: The proactive, personalized cultivation of new contributors was reported in multiple communities, whether formally or informally. Mentorship took many forms, from helping new contributors acquire wiki-skills, to introducing them to various wikispaces. In a few communities, this mentorship became an inculcated behavior, with no-longer-very-new contributors becoming mentors to newer contributors.
  • Off-wiki events: Casual gatherings for new contributors to edit together, in a relaxed and social atmosphere, were reported to help contributors gain confidence and become regulars. This form was reported as particularly useful in South, East, and Southeast Asia.

Example actions

edit
  • A study to document and compare good practices of new contributor engagement and growth, across different communities, to serve as a long-term knowledge base and point of reference for communities to consult or compare against.
  • Connecting contributors across communities with each other, to share information on implementing new practices
  • Discuss and possibly introduce new practices to interested communities
  • Support regular off-wiki low-barrier events, including funding (via WMF grants)

Community governance

edit

Community governance can include:

  • Decision-making processes and interactions, e.g. making decisions by consensus or by vote
  • Formal or informal policies, relating to both editing activity or user interaction
  • Formal or informal roles played by community members, e.g. sysops, committee members, and mediators

For this framework, the key capacities identified under community governance were: Roles, Policies, and Conflict management.

Roles

edit

Community members may play formal or informal roles within the community, to establish or enforce community policies and social norms. Unlike informal roles, formalized roles (e.g. sysops) will have various types, terms, and specific policies for election / appointment and removal.

Summary of Research

edit

Wikipedia Administrators (sysops) were the most commonly reported formal, on-wiki role (followed by Bureaucrats), with the number of administrators varying widely across communities; most interviewed communities had several dozen or fewer. The majority of these admins were elected by community vote (with rules such as requiring 70+% of the vote to be elected), with only a few communities reporting admins being appointed after public nomination.

Very little de-sysopping was reported across communities, with most happening on a case by case basis due to either inactivity or community grievance. Overall, communities did not report an expiration of adminship, with years of inactivity being the most commonly reported criterion for de-sysopping. Some communities reported difficulties with newbie-biting admins or burnt-out admins exercising the tools too aggressively.

Example actions

edit
  • A study to document and compare admin selection criteria, terms of office, and de-sysopping criteria across a diverse set of communities.
  • Document a few detailed case studies of change to those practices in a given community, with a before/after comparison.
  • Facilitate community conversation on changing policy in interested communities.

Policy creation and revision

edit

Policy creation and revision includes all activities involving the definition and revision of new or existing policies.

  • Y This includes the on-wiki spaces for policy discussion, and rules / norms for discussion and decision-making;
  •  N it does not include conflict management, which is a separate capacity.

Summary of Research

edit

Overall, communities reported that their policies have evolved over time, usually in response to community need, but that this continued evolution is a very involved and difficult process. They reported that while the assessment and reassessment of policies is a healthy practice, it is also sometimes too cumbersome, contentious, or exhausting, leading to demotivation of the community.

The policy-related issues reported include:

  • "Dead letter" policies, which are polices that exist on wiki but are not followed in day-to-day activities.
  • "Unsuitable" policies, which create continuous problems because they are not appropriate (or no longer appropriate) for the community’s context or needs, or the interpretation of the policy varies too much.
  • Policies that exist without clear implementation guidelines (e.g. policies with no clear way to request discussion or resolve disputes).

While the causes for these issues were largely community specific, one consistent theme that emerged was the practice of looking to other language Wikipedias (quite often English Wikipedia) for guidance on the definition or implementation of core policies (e.g. Five Pillars, Notability, Quality). While community members reported that this practice came with clear advantages, some communities also reported the downstream difficulties or dysfunction listed above.

The policies most frequently mentioned as being the hardest to implement or sustain were:

  • Notability, which was mentioned by almost every community with a number of associated issues, e.g.
    • Systemic bias, e.g. in discussing and determining "what is notable" based on local contexts
    • Verifiability of print sources that are not available to other community members
  • Quality, particularly the question of Featured Articles and whether that practice was suitable for their community. While most communities reported a current Featured Article process, a few adopted processes that were similar but with less stringent guidelines

Policies on deletion, language and neutrality / NPOV were also mentioned as critical policies, though core issues mentioned were largely community specific.

Example actions

edit
  • In-depth conversations with a wider set of interested communities about their current policies, to identify pain points and interest in change.
  • Given community interest in changing/improving/covering an identified policy area, do any of the following:
    • Compile a comparative policy review from a diverse set of communities
    • Facilitate a community conversation about policy change (e.g. setting timelines, "nutshelling", periodic recaps), if external facilitation is deemed helpful.
    • Design an experiment for a proposed policy creation/change, including a transition plan, measures of success, fallback procedure, etc.

Conflict management

edit

Conflict management emcompasses all the activities involved in raising, discussing, and resolving both inter-user and content-centered conflicts. It includes the various on-wiki spaces for discussing conflicts, as well as the formal and informal roles played by community members (e.g. mediators, admins, Arbitration Committees).

Summary of Research

edit

Conflicts were reported in every community, on topics primarily related to content (e.g. national politics, race, religion, caste, age), style (e.g. grammar, transliteration, language usage), or interpersonal issues. While very few communities reported set processes for raising conflicts, there were a number of processes and norms reported about conflict discussion and resolution:

  • Conflict discussion on village pumps. Many communities reported that conflicts tended to be resolved on article or user talk pages, with a few communities creating separate sections of the village pump or an entirely separate page for conflict discussion and resolution.
  • Conflict resolution via:
    • User blocks. Communities mentioned using an escalating series of blocks (e.g. one day, one week, one month) to discourage disruptive behavior or failure to comply with norms.
    • Committees, e.g. Arbitration Committees or Mediation Committees. Almost every community with a current or past Arbitration Committee also reported their limited use or effectiveness; communities without Arbitration Committees cited them being “too cumbersome” or time-consuming for their communities.
    • Appeals to more experienced editors. Every community reported different “generations” of editors, with many reporting that some members from the founding or “early” generations were still active within the community today. Those communities also reported the practice of appealing to those more experienced editors to weigh in on disputes, not to make a final decision but because they were respected members of the community.
    • Community vote. A few communities reported that very large conflicts were put to the community for resolution by vote.

Example actions

edit
  • Compile general (non-wiki) resources on conflict management, mediation, psychology of conflict resolution, etc., and make them available in multiple languages.
  • Compile wiki-specific resources on conflict management practices from other communities, and make them available in multiple languages.
  • Plan and fund local-language workshops for interested community members on conflict management by local professionals.

On-wiki technical skills

edit

On-wiki technical skills focuses on the resources, on-wiki spaces, and peer-to-peer mentorship/help available on wiki-related technical issues, e.g. creating/updating bots, graphics, templates, Lua programming, wiki markup.

Summary of Research

edit

The availability of community members with on-wiki technical skills varied greatly between communities. While some communities were able to identify a few (e.g. 1-3) community members that could help on bot-related issues, a few communities identified larger technical capacity. In these communities, dedicated on-wiki spaces existed for handling technical issues, e.g. a technical helpdesk, a graphics WikiProject, or a bot request page.

However, the need for technical capacity - on bots, template programming, graphics, and documentation - still surfaced across communities. Moreover, even for community members contributing to language projects with many existing technical tools, wide awareness of those tools and of how to use them remained a large obstacle.

Example actions

edit
  • In-depth conversation with interested communities on their current tech practices and tech needs, with an emphasis on automation, tracking, and evaluation needs (rather than on desired major features like a cross-wiki watchlist). Identification of these major tech needs would be followed by prioritization by community and evaluation within WMF.
    • If those needs can be addressed with existing solutions, focus on communicating, documenting, translating, or teaching relevant information about existing solutions.
    • For those sufficiently-impactful needs that are possible to address with a reasonable investment of effort, see if they can be worked into an existing WMF team's workplan (particularly via the new Community Tech team). If not possible, alternative actions could be: collaboration with an interested volunteer developer from the wider movement, development via an IEG.

Communications

edit

Communications encompasses sharing the Wikimedia story and purpose with a broader public audience. This can include: raising awareness, working with the press, pitching stories, building messages, drafting announcements, managing embargoed news, correcting inaccuracies, responding (and when not to respond) to negative coverage, leveraging social media and blogs, and more.

Summary of Research

edit

Awareness remains an issue for many communities, including awareness of the Wikimedia projects, but also awareness of “free knowledge” and an individual’s motivation and ability to contribute.

Most communities without a chapter reported little to no interaction with traditional media due to a range of reasons, e.g.

  • Traditional media companies that tend to cite English Wikipedia, rather than local language projects
  • Previous bad press or the need for anonymity motivating community members to actively avoid media relations
  • Limited presence of the local language in media overall
  • Lack of know-how

While many communities reported the use of social media (primarily Facebook, Twitter, and chat applications), with most communities these platforms were used more for community interaction, than for communications to the general public.

The communities who did report relations with traditional media all had chapters, though engagement still varied tremendously between proactive engagements (e.g. correcting mistakes, using media to promote activities) and reactive engagements (e.g. responding to an approach by a reporter interested in covering some aspect of Wikipedia).

Next steps

edit

The WMF Communications department is gathering information about movement-wide communication challenges and needs. If you’d like to get involved in sharing skills, solutions, or challenges please join here (link TBD).

Example actions:

  • Build a Communications portal, including curated collection of community good practices
  • Messaging resources, e.g. some frequently asked questions and answers for press & outreach
  • Plan and fund local-language workshops for interested community members on media relations by local professionals

Partnerships

edit

Partnerships include any relationship with non-Wikimedia organizations and individuals that are used to further the Wikimedia mission. This can include partnerships with:

  • Government
  • Cultural or academic institutions, e.g. GLAM, Education programs, or in conjunction with CIS-A2K in India
  • Individuals who donate their copyrighted content (e.g. authors, photographers), or otherwise contribute to the mission (e.g. providing space for Wikimedian activities; offering professional training to Wikimedia volunteers).

Summary of Research

edit

For communities that pursued external partnerships, there were four key points reported:

  1. Overcoming the general lack of awareness or negative perception of the Wikimedia projects was key in successfully establishing initial contact. As this initial contact could be made either through “warm introduction” (i.e. being introduced by a mutual contact) or through “cold contact” (i.e. reaching out to the partner formally, usually via their Outreach or Public Relations departments), assessing and addressing the existing level of awareness was a key first step.
  2. Understanding the priorities and focus areas of the potential partner was critical in creating specific projects that would become the basis for partnership.
  3. Setting expectations appropriately can be difficult, both from a cultural perspective and from a logistical perspective. Partners need to ultimately be comfortable with the way Wikimedia projects work and are governed (e.g. someone from another country could delete the photo you donated due to insufficient copyright information). Moreover, partners might expect more conventional project management terminology (e.g. budgets, “Memoranda of Understanding”) and disciplines (e.g. set timelines, deadlines, and “deliverables”), as distinct from the more natural “eventualism” of the work on the wiki.
  4. Overall, these partnerships are a long-term investment on the part of the community. Building the relationship to a point that is mutually beneficial to both parties can take months, if not years.

For communities that have worked with government agencies (at either the local, regional, or national levels), there were a range of different engagements and benefits reported:

  • Direct support or donation to the community.
  • Improved legitimacy and credibility through association.
  • Generally benefiting from government initiatives on open knowledge, language preservation and language projects more broadly (e.g. having a Unicode input method for the local language; using the products of a government digitization initiative).

However, as mentioned before, governments were reported to be either an obstacle or opportunity depending on the environmental context. The inability to establish partnerships or even avoidance of trying was also reported.

Example actions

edit
  • Outreach Wiki's typical scope.
    • Curate resources on good practices in partnership-building, by partner type and region.
    • Provide mentorship by experienced Wikimedians on how to pitch projects to partners, "how to get past 'no'", etc.
  • Plan and fund local project management training for interested communities by local professionals.

Notes

edit
  1. a b Interviews were conducted by Sati Houston and Asaf Bartov, with individuals from the following 16 communities / countries: Indonesia, Mexico, Philippines, South Africa, Turkey, Ukraine, India (English, Hindi, Malayalam, Marathi, Tamil, Telugu, Urdu) and the Arabic community.