When Pratik Shetty presented at this month’s global Wikimania conference in Stockholm, he told the story of Venky, the security guard at his apartment building in Mumbai, India. Shetty had noticed some mosquitoes in standing water nearby his building and alerted Venky that the insects could cause the tropical disease dengue fever. Venky responded that he could treat dengue by himself, without going to a doctor, and proudly showed Shetty a Hindi YouTube video that described five at-home remedies. Shetty told him that Wikipedia was a much more reliable source than YouTube and described how the information contained in the Wikipedia entry was supported by several third-party medical sources. To this, Venky replied, “Pratik, that’s all well and good, but I cannot read your Wikipedia.”
That sort of interaction is why he and co-creator Ian Furst launched a project called VideoWiki. At Wikimania, Shetty and Furst, an oral and maxillofacial surgeon based in Canada, made the case that it was high time Wikipedia dedicated more energy to video. Even in 2019, an estimated 750 million people around the world are illiterate, according to UNESCO. By limiting its encyclopedia to text and a few photographs, Wikipedia was failing to reach this large audience. The VideoWiki project provides web-based software for the collaborative creation, editing, and distribution of multimedia content on wikis. “Knowledge sharing can save lives,” Furst said. “And video is the only way in which much of the developing world can access that information.”
The medical information gap shown by Shetty’s security guard is distressing but not uncommon. YouTube is rife with popular but misleading medical videos. On the other hand, Wikipedia has repeatedly been shown to be a high quality and generally reliable source of medical information. Most of the videos released by the project so far are on health topics. Examples include video summaries of dengue fever, hypertension, cancer, and pneumonia, still a leading cause of childhood death. The videos can be created in English, Hindi, Arabic, French, Ukrainian, and Japanese.
These Wikipedia video summaries are not fun infotainment like Vox explainers, nor do they have the beautiful production value of Netflix’s Our Planet. In its current beta version, VideoWiki is more like Siri giving a PowerPoint presentation. Each “slide” is a free-use image or video, and the Google text-to-speech engine reads the wiki script for that section using a synthesized voice. For example, the video for major depressive disorder pairs Van Gogh’s oil painting Sorrowing Old Man (at Eternity’s Gate) with the text, “Major depressive disorder (MDD), also known simply as depression, is a mental disorder characterized by at least two weeks of low mood, that is present across most situations.” That’s not exactly groundbreaking information or punchy writing—but it’s accurate, sourced, and updatable. As the medical community learns more about causes of and treatment for major depressive disorder, the video will be amended to reflect the latest medical consensus.
Increasing the video content available on Wikipedia and its sister projects clearly aligns with global technology trends. A report from the Cisco telecommunications company predicted that by 2021, 80 percent of the world’s internet traffic will be video, up from 67 percent in 2016.
A Pearson study found that in the US, members of Generation Z, defined as ages 14 to 23, rank YouTube and video as preferred methods for learning by significant margins over millennials. And the market for internet video streaming in India is expected to grow from $500 million in 2018 to a potential $4.5 billion to $5 billion in 2023, according to a November 2018 report from the Boston Consulting Group.
Despite video’s exponential growth, Wikipedia is still predominantly a text-based encyclopedia. The Wikipedia community information page WP:Videos notes that “Wikipedia is still at an early stage in its process of incorporating videos into its encyclopedic content.” Prior efforts to incorporate video have been met with pushback. From 2015–18, the English encyclopedia piloted a partnership with medical education provider Osmosis to incorporate videos into articles. But once created, it was difficult for other Wikipedians to edit those videos, which goes against the ethos of the community. The Wikipedia-Osmosis partnership formally ended in May 2018.
Although the concept of an encyclopedia that “anyone can edit” was once considered revolutionary, it is fair to say that Wikipedia today has a strong traditionalist streak. A Dutch Wikipedian at the Wikimania conference described how several committees and stakeholders must sign off on even the smallest updates, like minor code enhancements to a bot that fixes broken URLs. “English Wikipedia is bureaucratic hell,” he told me. In a 2015 New York Times opinion piece, Andrew Lih noted that internal tensions and governance disputes between foundation employees and longtime editors were among the reasons Wikipedia had been slow to adopt a mobile-friendly editing environment, which risked alienating the younger contributor base. (Wikipedians at the recent Stockholm conference told me that the mobile interface had significantly improved.)
On the other hand, some of Wikipedia’s orthodox tendencies have proved invaluable. For example, the policy requiring that information be based on independent third-party sources (rather than first-person blogs or tweets) has helped cement Wikipedia’s reputation as the “good cop” of the internet throughout the misinformation crisis.
That dependability is exactly what Shetty and Furst want to bring to VideoWiki. As they presented at the conference, they were careful to explain how the new web-based video editor was based on the site’s existing norms and policies. Like any other wiki, the content is created collaboratively, with users having the ability to edit the script. The media is available on a free-use basis, since the photos and videos are imported from Wikimedia Commons. And citations for each statement are shown on screen as they are read, mirroring the notion of inline citations in encyclopaedic text. The message was that the new project is still Wikipedia—built on the values of free, publicly available, and accurate knowledge.
Wiki values came up in the Q&A portion of the presentation as well. If more people were going to YouTube rather than Wikipedia for health videos, one attendee asked, why not just upload the videos to that platform instead? “Because YouTube is at odds with everything we hold dear as a community,” Furst said. “It lacks consensus, it lacks editability. Somebody can take it, replicate it, and add a logo to it. It’s rife with charlatanism.” Shetty also noted that YouTube was overcrowded, with thousands of videos on a given topic whereas Wikipedia provides a single summary version per language.
But after the presentation, one Wikipedian told me the videos would likely end up on YouTube anyway, and further that many people in the community would not find this objectionable. Technically, anybody could distribute the videos after they were made so long as they complied with terms of the Creative Commons or applicable free-use license. The issue is that the YouTube copy would not necessarily be the latest version or reflect accurate metrics. Each time a new version was uploaded, it would lose all of the views associated with the prior version. That’s a problem because YouTube users tend to perceive videos with more views as being more trustworthy, even if view count stats can be highly misleading. If the VideoWiki project doesn’t take off, it could be because long-term Wikipedia editors, accustomed to text only, might feel uncomfortable in the new role of short film co-director. “We’re a community that loves communicating by text,” one Wikipedian told me. “Basically, we communicate exclusively by writing.”
But one running theme throughout the Wikimania conference in Stockholm was the distinction between specific Wiki projects, such as the English-language edition Wikipedia, versus the overall Wikimedia movement, meaning broad global action toward creating free knowledge resources online. It seems to me that this trees-vs.-forest mindset could be helpful to both VideoWiki and future augmented reality extensions of the encyclopaedia. If the goal is to provide every single person with the sum of all human knowledge, then Wikipedians should not allow fondness for one medium (text) to hold them back from their mission. At the same time, retaining key values from earlier projects will no doubt help guide Wikipedia’s future course. The overall idea is to continue as a living tradition, and to unleash more interesting videos in the process.