Strengthen the community/en

From Strategic Planning
Jump to: navigation, search

Summary Issue: Warning signs signal community decline

There are warning signs that the existing Wikimedia community is in decline, and new contributors are not replacing those who are leaving the projects.

The Wikimedia community has built a tremendous resource in an unprecedented organic open source collaborative process. The continued relevance and future of the Wikimedia projects, efforts to extend their reach and improve their quality, depend on a vital and strong Wikimedia community. The aspiration of the Wikimedia movement is to engage the world in sharing knowledge; considering its current state, the health of the community requires some care and attention.

There is much untapped potential growth for the community; many people use Wikimedia's projects but are not contributing to them. Further, the community has some key gaps; some segments of the population have a lot to contribute to the projects but are not well-represented in the community. Finally, there are clear signs that the community is shrinking and that this is happening across the board for Wikimedia projects--not just for the largest Wikipedias.

Enormous untapped potential for contributors to Wikimedia's projects

In both absolute and relative terms, the contributor base of the Wikimedia projects is smaller than one might hope or expect. For example, in March 2009, fewer than 50K English Wikipedians made 5 or more edits in the month.[1] In the same month, there were over 166M people who visited the English Wikipedia.[2] This means that less than 0.03 percent of visitors are actively contributing to the project. This trend appears to hold across multiple Wikipedias: for a given language Wikipedia, approximately 0.01-0.05 percent of readers are active or very active contributors.

Active contributors make up a small proportion of all visitors to Wikipedia.

Within the contributor base, a small minority of users (10 percent) make the vast majority of edits (90 percent), which suggests that the content that appears on Wikimedia projects is created by a relatively small group of individuals.[3] In comparison, social networking sites appear to have a more even distribution of editing: approximately 60 percent of users make 90 percent of contributions. This suggests that there may be an opportunity to "activate" more passive Wikipedians.

About 90% of contributions are made by 10% of contributors.

As noted by the Wikimedia Foundation's former Chief Program Officer, Jennifer Riggs, "My experience has validated what others have told me: [The projects have a] small group of vocal, active volunteers." Given how many people currently use the Wikimedia projects, and how few make frequent contributions, it appears that there is enormous untapped potential to engage more contributors to Wikimedia's projects.

Some groups are under-represented in the current community

Some demographic groups are noticeably absent from the Wikimedia community. In 2009, a survey of Wikipedia contributors in 21 languages confirmed that contributors to Wikimedia's projects are primarily males, age 18-30, who are college-educated.[4] Women are particularly under-represented.

Contributor demographics appear skewed toward particular characteristics.[5]

As Misiek Piskorski from the Harvard Business School described, "The way the technology is structured is deeply, deeply masculine and does not fit with the way that women work together." The culture of reversion, deletions, and editing wars may be less attractive to women, and this culture may be driving women's under-representation from the Wikimedia projects.

The underrepresentation of certain demographic groups, such as women and those over 30, suggests that there are some segments of the population who are less likely to contribute to the projects without some specific intervention or invitation. While it is difficult to know the exact impact of their absence from the projects, it is reasonable to assume that the quality and relevance of the content is diminished by their absence.

Other groups are less noticeably absent from the community; while no survey results provide data on the number of subject matter experts and academics who contribute to the projects, several of those interviewed suggested that this was an important gap on some of the projects. For example, our conversation with Frank Schulenburg from the Wikimedia Foundation suggested that this is a problem. From his perspective, the absence of individuals who have studied particular subjects through their academic and professional lives may be limiting the projects' progress toward increasing the sum of all knowledge--both in terms of breath (topics covered) and depth (quality of articles covering those topics).

Current community is in decline

Looking across all the Wikimedia projects, the number of contributors appears to be either stagnating or declining.

For example, 5-6 years after a Wikipedia’s founding, the number of contributors appear to peak.[6] This is true not only for the largest Wikipedias (in terms of article count), but also for smaller Wikipedias.

Active contributors appear to have peaked for many Wikipedias.

Many of the smaller Wikimedia projects are seeing similar trends; given that these projects have less than 1000 total individual active contributors, this decline is even more troubling.

Active contributors to other Wikimedia projects also appear to be peaking.

This trend appears to be driven by existing contributors' departure from the projects. A study of Wikipedia contributors[7] indicates that contributors turn over at relatively high rates: after 8 months of editing, contributors have a 50% likelihood of continuing their work. This suggests that the community constantly needs an influx of new contributors simply to stay at the same size.

Contributors becoming increasingly likely to leave Wikipedia as their time as an editor increases.

While there is no clear data on what is driving this trend, a number of those interviewed suggested two main drivers: technology and community norms. As Mitch Kapor of the Advisory Board said, "I think wiki mark-up is the problem. It’s not possible to have friendly tools and a more open community without different mark-up." Other Interviews suggested that the projects are becoming increasingly hostile and that contributors are relying on policies to revert and delete others' work.

The Palo Alto Research Center (PARC) recently released a body of research [8] that suggests that new contributors to Wikipedia (those making 1 edit per month) face significant barriers to contribution. New contributors see 25% of their edits reverted, whereas the most active contributors see less than 2% of their edits reverted.

Less active users see their edits reverted at significantly higher rates than more active users.

In addition, a current experiment by experienced Wikipedians logging in as new editors and attempting to recreate the usual learning curve of article creation and taxonomy or cultural mistakes for a new editor has found direct evidence that some over-enthusiasm at new page patrol is giving many new contributors the impression that we are an uninviting, norm-stringent, insular community.

The number of active admins on the English Wikipedia peaked in early 2008 and is currently 15% below peak. Graph by en:User:MBisanz

On the English Wikipedia adminship has become an increasingly difficult status to achieve, with far fewer admins appointed than is necessary to replace those who we lose or who cease to be active. The longterm trend is for a 1% monthly fall in the number of active admins.

The number of new editors peaked in early 2007, has oscillated widely but is if anything on a rising trend. Graph by en:User:MBisanz

Though the number of people creating accounts is less than in the first peak, it is still consistently over 200,000 a month, however only about one in 20,000 will go on to become an admin compared to more than one in 5,000 in 2006. The proportion of newbies who become admins has long been low, but we can console ourselves that our project still attracts an enormous influx of new users. However our ability to keep and develop new users has long been low and has deteriorated.

Task forces

Task force: Increase contributions from readers and under-represented groups

Many more people use Wikimedia's projects than contribute to them. To significantly expand Wikimedia's community, Wikimedia must be more effective at converting readers into contributors. This will require removing major barriers. A survey of Wikipedia readers from 2009[9] found that readers face two primary barriers that prevent them from converting to contributors: readers do not feel that their contributions would be valuable, and readers are unfamiliar with the technical process of editing.

Beyond removing barriers, the Wikimedia community may be able to do more to welcome newcomers. Interviews indicate that there is a lack of clearly-defined processes and help tutorials to welcome new users into the fold. Discussion on the Talk pages of this strategy Wiki and interviews with multiple individuals familiar with the Wikimedia community suggest the community has become less open over time. One potential example of this is that new users are finding their edits reverted at ten times the rate that the projects’ most experienced users are seeing their edits reverted.[10] These factors may all be contributing to the fact that the community is not being sufficiently replenished over time.

The Wikimedia projects rely on contributions from a broad range of contributors to ensure a broad range of content. At present, there appear to be at least two groups with high potential to add value that are not well-represented on Wikimedia projects:

  • Women
    • A 2009 survey[11] indicates that contributors are over 80 percent male. Interviews also indicate that the style of collaboration in a wiki-centric community (e.g., deletion of content, reversion of posts) may be more suited to men than women.
    • Given that approximately half of the world's population is female, this finding is troubling. This skewed gender balance may lead to content that is less relevant to women, and it may also limit the types of content created.
  • Subject-matter experts
    • Interviews indicate that subject-matter experts (such as academics) may be reticent to contribute to Wikimedia projects. Given the openness of Wikimedia projects, subject-matter experts may find their contributions edited, reverted, or overwritten by those less familiar with their area of expertise.
    • Given that academics, in particular, have other obligations on their plate, the possibility that their contributions will not be differentiated from those of non-experts may make them less likely to contribute.
  • Other high-potential groups. Our analysis identified two groups that are less well-represented on the Wikimedia projects that have high potential to add value to the projects. That said, we do not expect that this list is complete. There may well be other groups that have high potential to add value that do not currently contribute to the projects in high numbers.

See Increase contributions from readers and under-represented groups task force for the list of critical questions associated with this task force, as well as specific supporting materials.

Task force: Enhance community health and culture

Why is Wikimedia's contributor base declining? Many people believe that policy proliferation is a major driver of this trend, and that contributors are leaving because there are too many rules. Interviews confirm that the Wikimedia community is becoming increasingly reliant on policies to guide the editing process. Editing wars are a common occurrence, as is “Wikilawyering” (i.e., experienced contributors cite policies to justify the deletion or heavy editing of another’s contribution). Wikilawyering may be creating a less friendly community. Not only does this make the community a less collaborative space for existing contributors, it also discourages new editors from joining the fray. Moreover, reverts are on the rise for all contributors. [12]

While policy proliferation may be one driver of the high turnover rates within the community and slowing growth in new contributors, it is not likely the only driver. If policies are the "sticks" that enforce compliance with community norms, there likely also exist potential "carrots" that could incent high-quality contributions.

Turnover rates suggest that contributors may face burn-out or editing fatigue. There may opportunities to celebrate and better appreciate editors' contributions (e.g., rewards, awards) which could help increase retention.

See Task force:Community Health for the list of critical questions associated with this task force, as well as specific supporting materials.

Additional information and resources

Supporting analysis:

Interviews:

References

  1. Wikistats. http://meta.wikimedia.org/wiki/Wikistats
  2. ComScore. http://upload.wikimedia.org/wikipedia/meta/2/2a/ComScore_trend_data_on_WMF_Sites.pdf
  3. "New Twitter Research: Men Follow Men and Nobody Tweets." http://blogs.harvardbusiness.org/cs/2009/06/new_twitter_research_men_follo.html
  4. Wikipedia Survey – First Results. April 2009. http://upload.wikimedia.org/wikipedia/foundation/a/a7/Wikipedia_General_Survey-Overview_0.3.9.pdf
  5. Note that Russian respondents were excluded from this analysis because of their unexpectedly high response rate.
  6. WikiStats. http://meta.wikimedia.org/wiki/Wikistats
  7. "Wikipedia: A Quantitative Analysis." Felipe Ortega. http://libresoft.es/Members/jfelipe/thesis-wkp-quantanalysis
  8. "More details of changing editor resistance in Wikipedia." PARC. 7 August 2009. http://asc-parc.blogspot.com/2009/08/part-2-more-details-of-changing-editor.html
  9. Wikipedia Survey – First Results. April 2009. http://upload.wikimedia.org/wikipedia/foundation/a/a7/Wikipedia_General_Survey-Overview_0.3.9.pdf
  10. "More details of changing editor resistance." PARC. http://asc-parc.blogspot.com/2009/08/part-2-more-details-of-changing-editor.html
  11. Wikipedia Survey – First Results. April 2009. http://upload.wikimedia.org/wikipedia/foundation/a/a7/Wikipedia_General_Survey-Overview_0.3.9.pdf
  12. "More details of changing editor resistance." PARC. http://asc-parc.blogspot.com/2009/08/part-2-more-details-of-changing-editor.html