Wikipedia is one of the strongest survivors of the old web, and also one of the most obviously human-powered, thanks to a large number of editors who make changes around the world. But after celebrating the site’s 20th anniversary last year, the Wikimedia Foundation is turning to new — and more automated — tools in search of the next wave of contributors. To be adding designed features to help users make their own edits, including suggestions for easy first steps, such as linking different articles. It does this while trying not to weaken the ties of its individual communities — and, the Wikimedia team hopes, possibly even make them stronger.
Wikimedia has been testing additional features for newcomers since 2019 and is now officially announcing them at a platform-wide level. Users logging into Wikipedia accounts will see a landing page for new editors. They are assigned a mentor from a pool of more experienced site veterans who can answer questions. And through the landing page, they are urged to make minor changes, sometimes suggested by a Wikimedia-trained machine learning system.
“A lot of people would try to start editing but fail and don’t stick around.”
“The Wikimedia Foundation noticed that there were issues with retaining new editors, meaning many people would try to start editing but failed and didn’t stick around,” explains chief product manager Marshall Miller. The team began a research project in 2018 to test new methods of getting people to stay — first on relatively small wikis such as the Czech and Korean language versions of Wikipedia, then on larger ones, culminating in an English-language launch earlier this year.
According to Wikimedia’s surveys, most people start editing Wikipedia because they have a particular task in mind — such as writing a new article about something they’re interested in, contributing to an existing article, or correcting a typo. But they often don’t know how to get started, and the Wikipedia editorial community can be notoriously — to use a mild word — fussy. There are good reasons for this: the site is a widely trusted fact-checking source, and a high-quality bar helps keep it that way. But it means a lot of the initial edits are rejected, leaving people feeling like they’ve failed before they’ve even started. On large wikis, there is an ingrained set of rules that can make participation challenging, while on smaller wikis that don’t get as many visitors, there may be less obvious incentives to participate.
“The way we’ve thought about these features is kind of starting to say, it’s so hard to edit Wikipedia. There are so many barriers to entry. And there are two ways we can attack that. One was to say, ‘Let’s teach people how to do it.’ And so we’ve done some of that,” Miller says. “The other way was to say, ‘Editing Wikipedia is so hard. Let’s make easy ways to edit.’”
“With one thumb you can edit while holding the rail in the bus.”
Mentoring is part of that first attack route. The global Wikipedia community currently has 584 people signed up to mentor newcomers; the largest individual encyclopedia, the English Wikipedia, has 86. (About 122,000 accounts made a change to the English Wikipedia in the past month.) Mentors don’t work closely with every Wikipedian they’re assigned, but users are encouraged to email them with questions – many of which are quite simple, but can benefit of a one-on-one interaction with another person.
The second is to give newcomers a nudge in the direction simple edits they are more likely to make mistakes and suggest ways to participate. In addition to the standard editing tab, Wikimedia adds guidelines for suggested tasks for newcomers such as copying and an option called “structured tasks”, including things like adding relevant images and cross-wiki links to pages. A machine learning algorithm will suggest page images and links in topics new editors say they are interested in, and the editors can approve or reject them, acting as a human-level filter for an AI system. “These are some of the first edits you can do with one hand on your phone — like one thumb you can edit while holding the rail in the bus,” Miller says.
The accuracy of the algorithm itself isn’t exemplary: editors rate about 75 percent of link recommendations correct, and the number is between 65 and 80 percent for images, varying by wiki. But 90 percent of the edits people make with it to be preserve. The system is not yet available on English-language Wikipedia – it is still being trialled on smaller wikis – but Wikimedia plans to make it widely available eventually.
Wikimedia’s new system is designed to provide many of these interface-based rewards. For example, an “impact” section on the newcomers page lets people see how many page views the articles they’ve edited have received, giving them an idea of the difference they’re making. In tests, people who see the new features are about 16 percent more likely to make their first edit and — for people starting the process — 16 percent more likely to come back and make another.
If you’ve been using apps like Duolingo or Tinder, these little pushes might sound familiar. It’s a kind of gamification: a way to turn a difficult task into a series of small actions with symbolic rewards. These systems are also often criticized – described as “addictive” or manipulative.
“Part of our design is, how can the user realize they want to discover more?”
But the Wikimedia team sees his work as structurally different. For starters, there’s no real profit motive on Wikipedia – the goal isn’t to get people “addicted” to contributions, but to familiarize them with the process. For another, this work is done in public, with the results of individual trials and proposals documented online where the global editorial community can weigh in.
Some of the resulting discussions are high-level, while others are very specific to individual wikis. “They even help design the different algorithms for the different languages,” said lead designer Rita Ho. For example, the Vietnamese language Wikipedia had to adapt its algorithm to explain how the language defines the beginning and end of words. The administrators of an individual wiki can also choose to disable the features – although Ho and Miller say this has been rare so far.
While these changes are largely technical, the goal is to build the number of people who feel comfortable connecting with other people in the Wikipedia community, especially in smaller wikis that desperately need new editors. Systems like structured tasks should have people dip their toes in the water – but eventually they will have to jump in.
“There are community members who are concerned that the more newcomers interact with automated processes, the less they understand the fundamentals of the wiki process, the community-based process,” Miller acknowledges. “Because these communities, even though they need images and links, they also need their future administrators, their future people who discuss policy, the future people who write full articles of whole material. And so part of our design is, how can the user realize they want to explore more and dig deeper into this?”