Science and Wikipedia: a round-up

Entrance to the Wellcome Trust building, host of this year’s Wikipedia Science Conference

This post was written by Dr Martin Poulter, Wikimedia UK volunteer and convener of the Wikipedia Science Conference

The past year has been eventful and exciting for anyone interested in how Wikipedia can support the process and understanding of science. Here are a few stories that have caught my attention, plus a next step that anyone can take.

We knew that the free encyclopedia is one of the top ten most-visited web sites, but thanks to the charity CrossRef we now know that it is in the top ten sites via which people reach scholarly papers.

However, not all links are equal. A paper published on Arxiv and summarised in the MIT Technology Review finds that open-access papers are 47% more likely to be cited on English Wikipedia than closed-access papers. The authors concludeopen access policies have a tremendous impact on the diffusion of science to the broader general public through an intermediary like Wikipedia.”

Closed access versus open access can be a matter of life and death, as shown by a New York Times article about the African Ebola outbreak, which noted that some crucial research was practically unavailable because “downloading one of the papers would cost a physician [in Liberia] $45, about half a week’s salary.”

The single most-used source on Ebola in affected countries at the peak of the African outbreak was Wikipedia, as we now know thanks to a Journal of Medical Internet Research paper about Wikipedia’s medical content. The paper, summarised on the LSE Impact of Social Science blog, found that Wikipedia is now “the single most used website for health information globally”. The authors surveyed Wikipedia’s top contributors to medical content (those with more than 250 edits). Of 117 respondents, more than half were professionals in, or students of, healthcare.

If so many people are consulting Wikipedia for health information, the issue of quality becomes crucial. There is an active field of research comparing Wikipedia to other reference works. As even its logo makes clear, Wikipedia is a work in progress, and there are acknowledged weaknesses, but some scientific areas have reached an impressive standard. A paper published last year in PLoS One compares pharmacology in German and English Wikipedias against scholarly textbooks, concluding “Wikipedia is an accurate and comprehensive source of drug-related information for undergraduate medical education.”

It’s not just public understanding of science that is being shaped by Wikipedia, but even the publication process. A topic round-up on “Inferring Horizontal Gene Transfer” is the latest in a series of review papers published in both Plos Computational Biology and Wikipedia, providing both a fixed, citable reference and an evolving summary of current knowledge.

How to keep up with these rapid developments? There is no better way than attending the Wikipedia Science Conference this September 2-3 at the Wellcome Collection Conference Centre in London. Registration is just £29 for two days, including lunches. Geoffrey Bilder from CrossRef is amongst those talking about Wikipedia’s links to the scholarly literature. Speakers from University College London and Cancer Research UK will talk about improving and assessing Wikipedia articles about cancers. Daniel Mietchen will talk about new models of scholarly publication involving Wikipedia and Wikidata, and there is much more in a packed two days.

In an interview at Oxford University (video), well-known author Ben Goldacre argues that the current publication model for medical research “needs a kick up the bum” in the direction of openness. New research appears at a torrential rate that overwhelms any human reader, so we need results in a form that computers can easily query, gathering evidence from many studies at once. This requires the scientific community to tear down barriers to access, and that is what all the conference speakers are working on, in different capacities. Wikipedia, Wikidata and related projects are increasingly showing us what that transformed world of open science will look like.