
One of the perpetual tensions with efforts to fill knowledge gaps on Wikimedia projects is: How do you organize collaboration? When working with different topics, organizers and experienced on-wiki facilitators try to create more collaboration between editors, but because they are working on knowledge gaps, it’s a struggle. There would be no gap if a community was already collaborating to fill it.
For the last few years, the Campaigns team at the Wikimedia Foundation has been working on the CampaignEvents extension, with a focus on the online and offline events that drive most grant-funded collaboration. These programs have sizable influence, especially in growing communities in the Global South and with helping smaller Wikipedias and Wikimedia projects fill key knowledge gaps. But beyond events, the CampaignEvents extension has always had a focus on making collaboration easier more generally.
As we have worked on campaign organizer tools, however, we have been asking internally: Do WikiProjects and other permanent collaborations on-wiki need some of the same tools as grant-funded events? For example, do they need ways to track participants, track the work being done, and communicate between organizers and participants? What steps can we take to make the collaboration features from the CampaignEvents extension serve more permanent collaborations on-wiki?
What do we mean by a WikiProject and other permanent collaborations?
Efforts to build collaboration to fill knowledge gaps tend to fall at two ends of a spectrum:
- For newer audiences or audiences in the Global South, organizers tend to build off- and on-wiki campaigns. They need the social space, sense of timebound urgency, and off-wiki venues for collaboration. Though most of these activities include strong on-wiki community participation, the call to action and social component of a campaign or event creates interest, focus, and enthusiasm for the work that often includes off-wiki activities.
- For more experienced editors or members of language wikis where contributors have ample volunteer time on the internet, a permanent source of collaboration in the form of a WikiProject or task force can offer more structure. These structures often rely on a small core of motivated topical editors working on a shared purpose.

Whenever you attend Wikimedia events, WikiProjects (like Women in Red, WikiProject Military History, WikiProject Medicine, etc) get described as ideal spaces: places where collaboration and coordination is possible. We even have a non-trivial body of evidence that suggests that these are some of the most productive spaces on-wiki. When we first started exploring the topic, Wikidata recorded thousands of attempts to make WikiProjects work: English Wikipedia had over 2700 pages marked as WikiProject pages, with the next largest being frwiki (594), eswiki (462), zhwiki (358), ruwiki (296) and fawiki (280) (see Wikidata Query).

Yet, when you spot-check the activity of the WikiProjects, many of the projects are partially abandoned, never functioned, or are superficial attempts at collaboration. There is often a list of articles, a few registered users, and not much else. Even on English Wikipedia, 923 WikiProjects are thought to be active, 250 partially active, at least 171 are thought to be inactive and 686 defunct. It’s even very common for WikiProjects to get merged into bigger ones!
On most wikis, creating and maintaining WikiProjects with sufficient complexity to provide a place of collaboration is incredibly complicated. The structure of a WikiProject on larger wikis requires a lot of content and collaborators to keep active. Very few of the WikiProjects have evidence of newcomers being successfully welcomed. Meanwhile, on small wikis, the technologies that larger wikis developed for WikiProjects just don’t work; they often imply a lot of manual labor, bots, or infrastructure not available on the smaller wiki.
What can we learn from research?
To better understand how permanent collaborations were similar or different from events and campaigns, we did a two step research project between August and October 2024. The project included two phases: 1) an on-wiki survey (similar to the Clovermoss Survey) that allowed us to evaluate perceptions of effective collaborations like WikiProjects and 2) a survey (in English, French, Spanish, Arabic and Indonesian) focused on understanding how individuals understood collaboration on the wikis. All told: 25 people participated in the on-wiki discussion and 147 people completed the survey. You can read about the process on meta, and find the full report on Commons.
The highlights from the study include:
- Respondents tended to be very experienced (both by edit count and tenure), generally reporting participating in more than one wiki and placing high value on collaboration with other editors.
- English contributors reported more value from backlog drives than non-English contributors. Non-English contributors reported more value from off-wiki activities (workshops and editathons) and writing contests than English contributors.
- Most respondents said that they joined WikiProjects and other collaborations to work on content gaps and feel like part of a community.
- Most respondents indicated a preference for either a permanent collaboration or a few days of collaboration.
- The most common challenges identified by respondents relate to finding participants, engaging newcomers and keeping participants engaged in collaborations. This is consistent with user stories for Campaign Products tools.
- Non-English respondents were more concerned with problems related to infrastructure: creating projects, identifying content, and measuring impact.

The study highlights the big gaps between the larger language wikis, like English Wikipedia, where WikiProject infrastructure is well established and creates editor productivity, and smaller language wikis, which often do not have the building blocks to create a successful collaboration. Some of these gaps are the tools for creating and maintaining a WikiProject and ensuring that people find the project and feel welcome.
The results of the study helped us affirm that the tooling needs for WikiProjects and other on-wiki collaborations are similar to those already created by the Campaigns product team. The team will be able integrate this evidence into future work on the features of the CampaignEvents extension, in order to serve more kinds of collaboration in the future. For example, one feature they recently explored is adding other kinds of collaborations (like WikiProjects) in the second tab of the “Collaboration list”. The Collaboration list is promoted to people who register for events, with the idea that people who register for events could be looking for further collaboration opportunities.
At the same time, the study still leaves a large gap around our knowledge about how smaller language wikis build successful collaborations. Most of the examples shared with us in the two surveys were focused on larger wikis (mostly in the top 20 languages in terms of size) and in communities that are well integrated into the international community, with the non-English respondents of the survey often having experiences on 2 or more wikis. Moreover, in the written feedback, we had many comments pointing at the “WikiProject” as a permanent, everlasting collaboration not being sustainable. We are left asking questions like: Can we make a more agile, smaller collaboration that requires less work, especially when supporting smaller wikis?
What is already happening to improve WikiProject technologies?
WikiProjects are a sprawling set of social engineering, technologies, and data. Some wikis have the PageAssessment Extension, which allows wikis to track which pages belong to a WikiProject. Each local wiki has a family of templates and tools that are used for creating the WikiProjects themselves. And then users need to organize themselves to maintain the content list and tracking tools. When talking to editors across smaller language wikis, they often point to English Wikipedia and Wikidata as being at the forefront of these technologies, but unachievable locally. For example, the Ping WikiProject feature on Wikidata leverages WikiProjects to contact groups of collaborators, yet is hard to redeploy elsewhere.

Recently, several technical improvements have made it easier to form topical collaborations and make them more impactful:
- On Spanish Wikipedia, User:Sophivorous has been experimenting with various ways for less-active WikiProjects on smaller Wikipedias to form more rapidly, with a family of templates and javascripts that makes it easier to generate a WikiProject based on a list of articles and a clear scope. Not only does the tool generate a landing page for the WikiProject, but it allows you to introduce some light-to-newcomer backlog tasks similar to the Newcomer Growth tools but focused on the topics in the WikiProjects.
- Also, in one of the most recent updates to the search index, a new search special word allows search by WikiProject where the PageAssessment extensions is deployed. This of course has the caveat that the PageAssessment tool is only available on a handful of wikis.
- And, if you want to use lists created by other wikis to drive collaboration, especially around translation — the Language and Localization Product team built a tool that allows you to create a collaborative list, whichc an be promoted in the Translation tool to create a thematic target for translation. Now, international-multi-wiki campaigns can have priority lists easy to build and share with anyone focused on translation.
Yet these are incremental changes that don’t necessarily address the core concerns raised in the research, namely the need to find and retain participants in the collaborations.
Though the WikiProject as we know it might not be the solution for the future, in the face of an internet that is less focused on collaboration, and where fewer people sign up to edit Wikipedia, we need to lean into the opportunities for collaboration on shared interests and topics. Many editors arrive at Wikipedia focused on specific edits or articles but, if they are going to be Wikipedians, often it’s because they have a passion for free knowledge around a certain topic. In the face of a longer-term decline in new users on the wikis, safe thematic collaborations might be a way to retain the people who have already shown us they are interested.
Have any thoughts about on-wiki Collaboration or questions about the research? Give us feedback on the talk page on Meta.

Can you help us translate this article?
In order for this article to reach as many people as possible we would like your help. Can you translate this article to get the message out?
Start translation