Why Wikipedia Is Immune to Coronavirus

As the worldwide pandemic has been accompanied by an 'infodemic' of bad information, Wikipedia has stepped in to provide relief. So much so that it has become the go-to source for COVID-19 information

Omer Benjakob
Omer Benjakob
Send in e-mailSend in e-mail
Go to commentsכתוב תגובה
Firefighters wear protective face masks, amid fear of coronavirus disease (COVID-19), as they disinfect the streets ahead of the Iranian New Year holiday, in Tehran, Iran March 18, 2020.
Firefighters wear protective face masks, amid fear of coronavirus disease (COVID-19), as they disinfect the streets ahead of the Iranian New Year holiday, in Tehran, Iran March 18, 2020.Credit: WANA / Reuters

Last week, Wikipedia broke its record for most viewers per day, with over 360 million page views globally in one 24-hour period. Everyone is self-isolating and everyone is online – and Wikipedia is one of the beneficiaries. In Italy, where the virus hit especially hard, for example, many were using their down time to volunteer on Wikipedia and help enrich its local Italian-language content.

The COVID-19 pandemic traffic surge has brought with it special challenges to the web – from the struggle of Netflix to meet streaming demands, to the deluge of viral fake news relating to the virus – deemed an “infodemic” by the UN.

The misinformation ranges from false theories about the origins of the virus to suggestions to drink tea to lower fever, as well as claims that vaccines exist but are being withheld by dark forces. YouTube, Google, Twitter and Facebook have all taken drastic steps to curb the spread of such materials – warning readers against certain posts, and referring readers to the World Health Organization website for reliable information. They also are working with the WHO to take down content the latter deems dangerous. While efforts by the commercial online giants have been mostly successful – Wikipedia is not just doing its part, it’s actually having its moment. Since the start of the year, its English-language articles about the virus alone have registered over 115 million views, and in recent weeks alone, more than 40 million have read COVID-19 entries in all of Wikipedia’s languages.

In addition to breaking records, Wikipedia has received numerous fawning headlines for dealing masterfully with the proliferation of coronavirus information. The coronavirus crisis has led the community of volunteer editors and even the Wikimedia Foundation that supports them to acknowledge a role that has been thrust upon it, that of being the public’s main source of medical and health information. There are over 800 different articles dealing with detailing different aspects of the outbreak, from its spread in different countries to the ongoing race to develop search for a vaccine and even a piece about the xenophobia that it has caused.

Need a facemask? Why not just create one with your 3-D printer one from Wikipedia’s’ database of images (it even comes with folding instructions). Wondering about death rates in different countries? Wikipedia’s main table for coronavirus casualties updates itself hourly across languages, with data taken only from official sources. If that’s too hard to absorb, you can check out any one of Wikipedia’s graphic visualizations of data about the spread of coronavirus, based on material from its sister project Wikidata. Wikipedia’s educational arm, too, is working to create and share learning resources for cuts stuck at home, as schools around the world have been shuttered.

A 4th grader navigates his school's online learning system on a laptop at his home in Woodinville, Washington, U.S. March 11, 2020, amid the coronavirus pandemic.
A 4th grader navigates his school's online learning system on a laptop at his home in Woodinville, Washington, U.S. March 11, 2020, amid the coronavirus pandemic.Credit: Lindsey Wasson/ REUTERS

Indeed, while other websites refer readers to the WHO website on coronavirus, Wikipedia refers readers to, well, Wikipedia. The overall Wikipedia homepage now has a section dedicated solely to the COVID articles whose veracity they are confident about. Asked by Haaretz to gauge the merits of the English-language article on the virus, Prof. Nadav Davidovitch, who heads the department of public health at Ben-Gurion University of the Negev, said he was “pleasantly surprised by its quality.”

Elaborating, Davidovitch said, in a phone interview, that, “The article offers a relatively complex and nuanced picture of the outbreak of the coronavirus, its local, national and international ramifications, from more basic science to clinical, epidemiological, public health, to social, economic and political perspectives.” Davidovitch also praised the article’s “wide and diverse array of reliable sources – from medical information based on academic references such as the WHO, to trustworthy media outlets that provide good coverage of the more social aspects of the virus.”

The reason Wikipedia has done so well in the past few weeks is because it was prepared, having already acquired infodemic immunity. By the time COVID-19 first reared its head, in China late last year, Wikipedia had already experienced numerous public-health scares, including the SARS outbreak of 2003 and the Zika scare that began in 2015. With each new outbreak, it’s as if its editorial immune system kicks in.

Wikipedia’s immunity can be credited to people like Dr. James Heilman, a Canadian emergency room physician, and Shani Evenstein Sigalov, a member of the Wikimedia Foundation’s board of trustees, both of whom have led Wikimedia Medicine – a special group of editors across the globe focused on overseeing safeguarding Wikipedia’s medical and health articles. Heilman, who first set up the project, has been working frantically alongside other volunteers from WikiProject Medicine in recent weeks to maintain the COVID-19 articles, which fall under the project’s editorial auspices.

“Our editing community often concentrates on breaking news events in which content develops rapidly. The recent outbreak of the novel coronavirus has been no exception,” explains Heilman. Two main elements in Wikiproject Medicine’s response are deciding which articles to “lock” from public editing and maintaining a very rigid sourcing policy, so that only experienced editors familiar with the project’s editorial standards can actually contribute.

Generally, anyone can edit a Wikipedia article. The exceptions are articles about contentious subjects (such as Israel/Palestine) or about topics in the news, which can be locked to anonymous editing so as to prevent vandalism, if enough registered editors vote for such a move. Locking an article, says Heilman, “is standard during these sorts of events [medical scares], as some level of vandalism is inevitable. Standard mechanisms however allow the community to rapidly correct inaccuracies and unsourced or poorly sourced content. Our emphasis is on not only using, but actually requiring high-quality sources to allow us to rapidly remove inaccuracies.”

A man wears a face masks to protect against the coronavirus as he queues at a supermarket in San Juan, Puerto Rico on April 7, 2020.
A man wears a face masks to protect against the coronavirus as he queues at a supermarket in San Juan, Puerto Rico on April 7, 2020. Credit: AFP

Evenstein Sigalov has spent the past half-decade training medical students in Tel Aviv to write and edit for Wikipedia. Her course was the first of its kind in the world, and her students have gone on to write over 10 percent of all the medical content on Hebrew Wikipedia. She explains just how rigid Wikipedia’s sourcing requirements are on medical content. “While most Wikipedia articles just require a source, medical and health articles have a higher standard – requiring secondary sources based on meta-analysis of multiple studies.” To that end, the project has forged partnerships with the WHO and with Cochrane, a database that summarizes the results of medical research. The latter, she says, “creates the secondary sources we need, and which, as far as Wikipedia is concerned, are better than single-case studies.”

Dr. Heilman explains how this sourcing policy is key for weeding out bogus information: “For example one idea is that [coronavirus] spread is related to the Australian bush fires. No decent reference was available, and therefore it was not added,” he says. There was also a discussion “about ‘HIV inserts’ within the virus and about suspicions that it may be man-made. But the source was a preprint [an academic article not yet subjected to peer review]. Not sufficient for inclusion and thus removed. Was a cure found by Thai researchers? We need a better source.”

One interesting facet of Wikipedia’s response has been the role of non-medical experts in fighting the spread of disinformation. According to Evenstein Sigalov, “some of our editors are doctors, some are med students or researchers, but some are just Wikipedians. That is the beauty of Wikipedia, anyone can contribute, you just need to stick to reliable sources.”

Marielle Volz, a volunteer with the project who has been involved with the COVID-19 articles, exemplifies exactly how this works and how even non-medical experts can play a huge role by only enforcing editorial standards. ”The biggest issue is that as with any new or emerging virus or event, early information tends to be especially unreliable – even sometimes information from scientists! The first paper that was published about the origin of the virus suggested it came from snakes... this paper was met with skepticism from the scientific community and this was reflected in the page with only a slight delay,” she says proudly.

Over the past few weeks, the community has even set up a special project that offers an organized list of sources on COVID-19 editors can use. One of its first tasks is translating the high-quality English articles into other languages. A similar project has been set up in India to make sure only accurate information is being spread across the globe. The project also monitors social media to see what articles are being shared and works to make sure they too are up to speed.

“Keeping Wikipedia reliable and up-to-date involves deleting material just as much as adding it, Heilman summarizes. “It appears that Facebook and Twitter have no mechanism to suppress false information rapidly or at scale – even in English,” he says. Victor Grigas, another volunteer, sums up the comparison to social media aptly: “Wikipedia has a bullshit detector. Wikipedia is run by volunteers who like to be volunteer police. It’s built -into the thing. Facebook just doesn’t have this.”

Comments