The Wikipedia machine relentlessly churns out information over which women struggle to have any influence. Photo: Jonathan McHugh/NS

Wikipedia is “like a sausage”, its founder, Jimmy Wales, told a reporter in 2004. “You might like the taste of it, but you don’t necessarily want to see how it’s made.” Back then, the free online encyclopaedia that anyone can edit was an exciting new, scrappy, collaborative utopia. Now it is the most influential source of information in the world. Wikipedia is often the first search result when we google something, our first destination when we want to understand something, and the place where academics, journalists and politicians first brief themselves, even though they might pretend it is not.

Dismissed as dangerously unreliable in its early days, Wikipedia has become more rigorous over the years, with references essential to the survival of any article. We trust the website much more: amid the early panic of the ebola outbreak, the Wikipedia page for the virus was seen as an authoritative, reliable source, receiving as many hits as the World Health Organisation’s online ebola fact sheet. Wikipedia has become one of the most recognised brands in the world and for many people it is the portal to knowledge in the 21st century.

Yet when it comes to how it is made, Wikipedia is a colossal failure. Only a tiny proportion of users now edit articles and the overwhelming majority of those editors are male. The most recent survey by the Wikimedia Foundation, the charity that supports but does not control Wikipedia, found that 91 per cent of the editors are men. More optimistic surveys have put the figure at 84 per cent – but still, Wikipedia has a huge diversity problem. Instead of being the egalitarian “sum of all human knowledge”, as Wales had originally hoped, the English version of Wikipedia is mostly the sum of male knowledge.

The gender disparity has skewed the encyclopaedia’s content – not only which pages are created but also which ones are worked on and improved so that they reach a high standard. Take its “List of Pornographic Actresses”; it is meticulously referenced, with clear sections according to decade. The page is organised, clean and easy to use. Compare it to the “List of Female Poets”: a sprawling dumping ground, organised by name rather than date, unreferenced and of little use to anyone unless they want to know whose name might come after Sylvia Plath in an enormous alphabetical list. The list of poets has been edited 600 times, by nearly 300 editors. The list of female porn stars is a newer page but over 1,000 editors have edited it more than 2,500 times.

Female poets at least get their own list. In areas such as science and technology, women are severely under-represented. If there is not a decent biography of a given woman on Wikipedia, users will assume she cannot be notable because she doesn’t have a proper Wikipedia page, so the marginalisation becomes circular and self-perpetuating. The biographies that do exist often put a woman’s status as a wife, mother or daughter in the first paragraph, before or next to her notable achievements. These personal details are more often an afterthought in biographies of men. Conventionally female interests are also neglected: there’s a single page for all six series of Sex and the City, whereas there are 43 separate articles on Top Gear. And when it comes to articles on topics such as rape and abortion, the gender gap among editors really begins to matter.

Wikipedia knows this is a problem – there is even a Wikipedia article on the subject (“Gender bias on Wikipedia”) – but no one knows what to do about it. Sue Gardner, a former executive director of the Wikimedia Foundation, set a goal in 2011 to increase the proportion of female editors to 25 per cent in four years. Just before she left her post in 2014 she confessed that she had not cracked the problem. “I didn’t solve it. We didn’t solve it. The Wikimedia Foundation didn’t solve it,” she said.

At the annual Wikimania convention in London last August, Jimmy Wales said the organisation had “completely failed” in its attempts to increase women’s participation drastically. “We’re really doubling our efforts now,” he said. “We didn’t do enough. There are a lot of things that need to happen to get from 10 per cent to 25 per cent: a lot of outreach, a lot of software changes.”

***

Elsewhere on the internet, women outnumber men on some of the other most visited sites, such as Facebook and Twitter, and in many online games. Why do they feel less welcome on Wikipedia? “I don’t want to get into a fight on the internet. Ugh,” says Zara Rahman, 26, originally from Man­chester and now living in Berlin. She trains journalists to use data and technology, so you might expect her to feel at home on Wikipedia. But her experience there left her “really annoyed. Just exhausted.”

The frustration stemmed from her experience editing the online entry for Hedy Lamarr, a 1940s Hollywood star and long-neglected inventor. Lamarr devised a crucial technique that paved the way for wireless communication, but her scientific achievements had barely a mention on her Wikipedia page when Rahman first looked her up. She edited the article to reflect the significance of Lamarr’s invention, referencing it in the first paragraph, but her changes were quickly reversed by another editor, on the grounds that Lamarr’s acting career was more noted by historical sources than her invention. Then someone added a line to the opening paragraph about how a film director had once commented on Lamarr’s “strikingly dark exotic looks”. The editing community allowed that to stay in.

“The page is actually worse than when I first found it,” Rahman says. “As it currently stands, a comment by a man about her appearance is more important than the fact that she basically invented wifi.” Lamarr’s invention is mentioned “something like three screens down. If you were looking for quick headlines about this woman, you’re going to stop at the fact that she appeared nude in a scene. That’s all you’re going to remember about Hedy Lamarr.” Sources matter on Wikipedia – the more references a fact has to back it up, the more likely it is to remain on a page – but that can lead to a systemic bias. “Of course her [Lamarr’s] acting career appears in more sources,” Rahman says. “She was a woman in the 1940s, there were men writing, and the men were writing about her being beautiful and exotic, not about women contributing to science.”

Rahman had dabbled in editing before she arrived at Lamarr, but after this encounter she stopped. “I wanted to edit because it’s fun and I think it’s important, but a Wikipedia editing war is not my style,” she says. Editors can be notoriously brusque, sometimes forgetting social niceties when they change other people’s work. The internet is littered with the blogs of bitter ex-Wikipedians who have been burned by rejection and the often fraught arbitration process the encyclopaedia uses to resolve disputes. Plus, Rahman was aware that she had hardly any clout, in Wikipedia terms, because she had not edited much before.

The conflict and hierarchy specific to Wikipedia may have been dispiriting but it was an internet-wide problem that ultimately put her off. “I’ve seen so many women be trolled and abused online, I don’t even want to dip my toes into that,” Rahman tells me. “I use the same Wikipedia name as I do for my Twitter and my blogs. If things are going to get vicious, it would be very easy for someone to find where I work as well as my email address.”

It is not just new users who feel alienated – even women such as Theresa Knott, who has been editing Wikipedia since its launch in 2001, have stopped contributing. She was once a leading figure on the encyclopaedia, elected to administrator and then arbitrator status, a role akin to that of a high court judge. But gradually she lost interest and she last edited in 2012.

“When Wikipedia was smaller it was a very different beast,” Knott tells me when we meet near the London mixed independent primary school where she teaches science and computing. “I met a lot of people and had great discussions in the early days. I wasn’t drawn to it because of the community but I stayed because of the community.

“Now editing is more of a solitary thing than it used to be because Wikipedia’s so much bigger. I think women like group activities more than men do; women like to socialise, and because it’s bigger I suspect it’s less appealing to women than it used to be.” When the community was smaller it was more collaborative. Editors took time to help each other learn the ropes, Knott says. “Now, it’s got very formal. I feel sorry for people whose articles aren’t the minimum length and don’t have at least one reference in them, because they just get deleted. That would put me off editing in the first place.”

It is hard to know how the gender gap has changed over time – the earliest survey of editors wasn’t carried out until 2010, when Wikipedia was already nine years old – but Knott says there were always many more male editors. “The women who were on there were more likely to be people like me rather than people with interest in . . .” – there’s a long pause while she searches for the appropriate words – “typical women things.” What does she mean by women like her? “Very geeky kinds of females who thought in a certain way and kind of fitted in with the men. There weren’t many women who would not traditionally be in a male sphere. When I did my physics degree, the ratio was 6:1. You kind of get used to it.”

***

If you’ve ever clicked on the “Edit” tab on a Wikipedia article, you will understand that having a particular kind of conventionally male-brained thinking might help on Wikipedia. Reams of code cascade down the page: curved, square and curly brackets, chevrons and underscores. It looks more like a computer program than a draft of an encyclopaedia entry. If you can see past the symbols to the bit of text you want to edit, it becomes straightforward: you put your cursor in the place you want to make a change and then type, or delete. Then you write an edit summary describing your changes and click Save – though there is no guarantee they will stay. Most edits, particularly changes from new users, will be scrutinised by an army of experienced volunteers and Wikipedia robots, looking out for mistakes, vandalism, libel and things that break the site’s code of practice.

Knott has observed a gender disparity among her young computing students: the boys have embraced coding more wholeheartedly than the girls, and are more willing to do it on their own, outside class. Even if Wikipedia didn’t exist, the highest-ranked pages on Google would still be more likely to have been created by men than women, she says. “It’s not just a Wikipedia thing – it’s an internet thing.” Wikipedia is about creating content rather than websites but all the behind-the-scenes manoeuvres that go into creating a page mean it has more in common with coding than editing a Facebook status, where the social network invites you to share “what’s on your mind”.

If there are going to be more female editors, Wikipedia needs to learn from websites where women feel comfortable. Some believe Wikipedia “editathons” might be the answer, where editors meet in person to work on neglected topics together. These are encouraged and supported by the Wikimedia Foundation, which sometimes provides tea, biscuits, laptops and trainers to help new editors learn the craft. Recent editathons have focused on topics such as ballet, Australian female neuroscientists and women in Jewish history.

While increasing the coverage of women on the site, these meet-ups are also more likely to attract female editors in the first place. Claire Millington made her first edit at a “Women in Archaeology” editathon in 2013. We meet at a café next to Senate House Library, where she has been working on her classics PhD at King’s College London. Her thesis is on the women who served in the households of Roman auxiliary army commanders, a group of women that has never been systematically studied. “There’s a pattern in what’s written about women and their achievements, and it’s basically that they’re not written about,” she says. “I don’t want Wikipedia to be a place where women are written out of history again, because if it’s not on Wikipedia, it’s not visible.”

Millington sees it as her duty to make sure that her academic field is properly represented on Wikipedia. She creates new articles and nurtures them, keeping them on a watchlist so that she can check on new contributions. So far, she has not yet found any edits that she’s wanted to change. Wikipedia’s genteel classics pages are unlikely sites for bitter editing wars, but Millington has yet to experience the encyclopaedia’s aggressive side, and has organised her own editathon, encouraging her colleagues to participate.

“I think the interface is the one thing that Wikipedia, Wikimedia, really needs to address. It’s not immediately intuitive,” she says. “It’s great if you’re techy – and there are a lot of people involved in Wikipedia who are techy – but the majority of the population are used to getting their phone out of the box and turning it on and using it. It’s not that women can’t do it, it’s just initially it’s not very welcoming.”

Is there another reason why women are less willing than men to contribute to Wikipedia: that women like to feel they have comprehensive knowledge of something, backed up by evidence, before they claim to have the authority to comment on it, whereas men are more prepared to blag? It takes confidence to believe you have the right to write an encyclopaedia entry, something men might have in greater quantities.

“[That’s] not really plausible,” says Charles Matthews, a former Cambridge academic and one of Wikipedia’s most prolific editors, when I put this to him. “To the extent that women have a different working pattern, they are more likely to be patient writers, that’s all. And motivated by different considerations.” The idea of different working patterns has come up before as an explanation of the gender disparity, in another way: several studies have found that women have less free time than men to dedicate to projects such as Wikipedia because they do more of the childcare and housework.

For Matthews, maybe the gender gap is being blown out of proportion. “There are other, similar systemic issues that are also important. Do Hollywood films get better coverage on Wikipedia than Bollywood? You bet,” he says. “We’re beginning to think there’s less of a gap in terms of writing rather than tech maintenance work on the site – which is lost if you treat all edits as equal.”

I can’t help thinking that if women were more confident about asserting their knowledge, they’d feel more at home on Wikipedia. Roberta Wedge, a former gender gap project worker for Wikimedia UK, agrees. “I think far fewer women would describe themselves as experts than men, but you don’t need to be an expert to edit Wikipedia. And there are many ways of contributing, like photography, like labelling and categorising things. Like adding links between articles so that when you’ve found an amazing, obscure woman you can make sure the article can be found from other places.”

Wikimedia UK hired Wedge for four months last year to address the gender disparity. She helped with editathons and attended related conferences. As she told me while she was still in the post, “My job is to say: there are fascinating women out there on the historic record, we need to get them reflected on Wikipedia, and men and women can add to that.”

The focus seems to be on making sure “female” subjects and women’s biographies are adequately represented, rather than recruiting women to edit, but the hope is that once those topics are better represented, ­female editors will feel more welcome.

But there is a limit to what the international Wikimedia Foundation can do. It’s a charity: there is no army of engineers who can make the editing interface more friendly, no funding for focus groups to reveal what women want from Wikipedia. Any intervention beyond that would undermine what makes Wikipedia great: the fact that it is built from the ground up, a collaboration that polices itself. The answer to the problem has to come from within Wikipedia. Ideas from the site’s discussion boards include a Girl Scout achievement badge in Wikipedia, and persuading celebrities such as Oprah Winfrey to ask their audiences to try editing. But ultimately it is up to women to choose to get involved, and up to existing contributors to make them feel welcome.

After several months away from Wikipedia, Zara Rahman met Wedge at a conference, and Wedge persuaded her to give it another try. Rahman has made a few additions to the biography of Marie Tharp, an oceanographer who created the first scientific map of the ocean floor. But she still sounds badly bruised by her experiences on Wikipedia, and is wary of becoming more involved. I ask if she even uses the site for reference any more. “Of course,” she laughs. “Where else do you get your information from?”