Wikipedia at 15: The sum of human knowledge?

Martin Poulter, Wikimedian in residence at Oxford University, considers the active encyclopedia’s first 15 years.

In 2016, Wikipedia reached fifteen years of age. It also celebrated making thirty million educational images freely available. In December 2015, its community of writers collectively earned the Erasmus Prize, an award for outstanding intellectual contributions to culture and society. With very much less financial power than conventional media organisations, it has become a towering giant in terms of cultural impact.

Wikipedia is part of Wikimedia; a family of web sites whose goal is that “every single person can freely share in the sum of all human knowledge.” Unlike other popular sites, Wikimedia exists purely for this purpose. It has no shareholders, no sponsors, and does not trade in its users’ data, as many other popular sites do. Each Wikimedia site is multilingual and freely accessible. As well as Wikipedia, these include Wikimedia Commons (a database of images), Wikisource (a library of out-of-copyright books and documents), and other sites supporting education and reference.

The goal of free knowledge for everyone is ambitious. What has been achieved in fifteen years is only a start. Still, to hundreds of millions of people who are information-poor, it makes a huge difference. Even people without an internet connection are benefitting: the charity SOS Children’s Villages sends Wikipedia on computer drives to schools in remote areas of Africa and Asia, while the Kiwix project gives people tools to access Wikipedia and similar sites if censorship or lack of technology prevents them reaching the internet.

Compared to the ambitious standard it sets itself—an encyclopedia for the world, written by the world—Wikipedia’s present state falls far short. Its volunteer authors are overwhelmingly male, white, childless, and live in Northern Europe or North America. So, as useful as it is, we have to be aware of its origins and its inevitable bias. Some of this bias comes from Wikipedia itself, but much of it comes from the world outside Wikipedia.

Bias on Wikipedia

Wikipedia’s policies require that articles are written in a balanced way, reflecting the points of view in the best available sources. Enforcement of these policies is up to the community: Wikipedians police each other. In a very active community, such as the English or German versions, a controversial article many be “watched” by dozens of editors. It is hard for a someone to use such an article to push an agenda. In a smaller community, there is less monitoring and more scope for articles that promote a point of view. For instance the Croatian version, with fewer than 500 active users, has had problems with articles that promote far-right politics.

Even when each article is written in an entirely neutral way, there is also a more subtle kind of bias to do with the emphasis given to different kinds of information. People write about what they are interested in, and what they can easily find information about. They naturally prefer to write about places they visit, entertainment they enjoy, or subjects they studied at school or university. So Wikipedia’s authors are more likely to write detailed articles about a village in England than about a city in Africa; more likely to write about a famous Hollywood actress than about the various kinds of discrimination faced by women. Of course there are many of us working to fill the gaps, but real progress requires more real diversity of contributors.

In the list of Wikipedias by number of articles, North-European languages, such as English, German, French and Dutch, dominate. Chinese and Arabic, each spoken by more than a billion people, are at positions 15 and 20 in the list. Researchers at the Oxford Internet Institute have studied geographic inequality, showing that there are more articles about the Netherlands (population 17 million) than the whole of Africa (1.2 billion).

The existence and length of Wikipedia articles can reinforce cultural prejudices about which achievements, or which aspects of human history, are important. It is hard to measure objectively how this subtle bias affects broader culture, but we know the direction of the bias. The most commonly accessed source of free information reflects the interests and needs of a European man much more than those of a woman in the Middle East. It favours men who have distinguished themselves as military or political leaders, and women who are famous for being attractive to men.

Giving the world the Edit button

Wikipedia grows by a process of gradual, collaborative improvement. Someone clicks the “Edit” button at the top of the article, makes a change on their computer or mobile device, then clicks “Save page” to publish that change. There have been nearly three billion of these “edits”. Sometimes people deface articles or spread hoaxes, but they are outweighed by edits that fix, improve and add knowledge. A single article can result from tens of thousands of edits, from more than a thousand different people, and can reach thousands of readers each day.

This is potentially the most world-changing feature of Wikipedia: not just giving the world a free encyclopedia, but giving them the Edit button. When Adama Diop, a student in Dakar, Senegal, searched for information about his home village of Agnam-Goly, he was told there was no article about the village, but that he could create one. He took the invitation, wrote a (French language) article, and even saved money for a digital camera to share photographs of life in the village. Other sources of knowledge have their advantages, but only Wikipedia offered Diop this opportunity.

The Edit button does not know or care who you are or where you come from. Yet in reality there are many factors—technological, cultural, and personal—which encourage or discourage people from clicking the button. Wikipedia promotes itself as a site that “anyone can edit” but that is only true in the strictest sense.

Cultural factors include spare time and education. Whether someone writes for Wikipedia depends on whether they have spare time and also their own access to knowledge: whether they went to university, and whether they have books in their home. Cultural expectations affect how much free time people have. For instance, if wives rather than husbands are expected to care for the home and family, women have less opportunity to volunteer for a project like Wikipedia.

Writing for Wikipedia involves a mix of reading, writing, and comparing. In practice, it needs a computer with a large screen and multiple applications. So it is easier for someone with a desktop computer than for someone whose Internet access involves a mobile phone. The Wikimedia Foundation and volunteer programmers have made heroic efforts to ease mobile editing, but only so much can be done on a phone.

The ability to contribute to Wikipedia depends not just on the device but on the internet connection: it is hard without a connection that is fast and reliable. Unsurprisingly, the global inequalities in Wikipedia articles closely resemble global inequalities in broadband accessibility.

Broadband Affordability courtesy of Mark Graham and Stefano De Sabbata, Internet Geographies at the Oxford Internet Institute. (Creative Commons Attribution NonCommercial) Source: http://bit.ly/1AdD0No
Broadband Affordability courtesy of Mark Graham and Stefano De Sabbata, Internet Geographies at the Oxford Internet Institute. (Creative Commons Attribution NonCommercial) Source: http://bit.ly/1AdD0No

For greater equality, there would have to be big changes in how people across the world access the internet. So for the foreseeable future, Wikipedia will be writable for some parts of the world and mostly read-only for others.

Access to knowledge

Wikipedia does not do its own research, but only summarises what is produced by the existing system of research and publication. If this system were equally open to everyone in the world, that would promote an equal opportunity to improve Wikipedia. Instead, access to published research is grossly unequal; more so than Wikipedia itself.

A lot of scholarly research is owned by European and North American publishers, who have a financial interest in restricting access. Scholarly research is also overwhelmingly published in English. So I, working in a university in the UK, have no problems reading the latest research, while other people face multiple barriers, even though their need is more urgent. An example of this, discussed in the New York Times, is the 2015 Ebola outbreak in the African country of Liberia. Medical authorities in Liberia were slow to respond because it was hard for them to access the results of medical research conducted in their own country:

“The investigators collected their samples, returned home and published the startling results in European medical journals. […] Even today, downloading one of the papers would cost a physician here $45, about half a week’s salary.”

I mentioned above that there are more Wikipedia articles about the Netherlands than the whole of Africa. The same researchers found even greater inequalities in scholarly publication. Four times as many academic papers are published in the Netherlands as are published in the whole of Africa.

We must also consider active censorship. Wikipedia’s mission is to share the sum of all human knowledge. This has to include knowledge about: political systems, including democracy; religious beliefs, including atheism; sexuality, including homosexuality and contraception.

Not everybody welcomes this gift. Around the world, there is political and cultural resistance to the free availability of at least some of this information. Censorship can affect Wikipedia itself or the sources that Wikipedia is written from. Wikipedia does not censor at the request of any government, but several governments have prevented their citizens from reading all or part of Wikipedia.

The good news

Wikipedia is written and edited by volunteers, but supported by the Wikimedia organisations, which include the Wikimedia Foundation in San Francisco as well as national or local groups around the world. These organisations put a lot of effort into greater diversity of content and of contributors. As a result, the software, the instructions, and the process for requesting help—all of which can seem intimidating for new users—are being continually improved.

Increasingly, we have the support of powerful friends. A growing number of universities, libraries, museums and other institutions are working in partnership with Wikimedia to improve freely-available knowledge. In my own work at the Bodleian Libraries in Oxford, I have been sharing images from the libraries’ collection of non-Western art and texts, to help improve articles about other cultures. I and my colleagues across the world run events to address Wikipedia’s gender bias, in which we create articles about women and their achievements. In the process we train new contributors, including more women than men.

Geometric diagrams from an 11th century Arabic manuscript, courtesy of the Bodleian Libraries, Oxford. (Creative Commons Attribution) Source: http://bit.ly/2dFygvo
Geometric diagrams from an 11th century Arabic manuscript, courtesy of the Bodleian Libraries, Oxford. (Creative Commons Attribution) Source: http://bit.ly/2dFygvo

A lot of effort is going into tools to collect and share facts and figures such as the dates and locations of historic events. The newest Wikimedia project is Wikidata, a multilingual database with many millions of these facts. This and other innovations make it easier to create educational materials in more topics, for more languages. Unlike an article, a database can build maps, timelines, and applications for people to interact with, not just read.

It is exciting that we will have more ways to access knowledge than via written articles, yet these developments can introduce their own bias. Not all knowledge can be summed up by numbers in a database. Some of the knowledge most worth sharing is about political, social, and cultural issues (free expression, for example) where different perspectives need to be explained with carefully chosen text.

To make knowledge freely available is a great opportunity, but success depends on people taking part. Wikipedia, its related projects, and the open internet that allows them to flourish should be cherished by anyone who can benefit from sharing the world’s knowledge. That means all of us.

Martin Poulter is Wikimedian in Residence at Oxford University.

Read more:


Comments (1)

Automated machine translations are provided by Google Translate. They should give you a rough idea of what the contributor has said, but cannot be relied on to give an accurate, nuanced translation. Please read them with this in mind.

  1. How interesting!

Leave a comment in any language

Read More

Swipe left to browse all of the highlights.

Do you agree with this principle?

Yes No


Free Speech Debate is a research project of the Dahrendorf Programme for the Study of Freedom at St Antony's College in the University of Oxford. www.freespeechdebate.ox.ac.uk

The University of Oxford