There are two parts to the story of the birth of the internet. Networked computers exchanging packets of information have existed since the 1960s, facilitated largely by a government project in the USA. The interface that allows anyone to share anything with anyone, anywhere – the combination of browser, webpage and, crucially, hyperlinked text that we use today – came later. Famously, the idea came from a paper on information management written by Sir Tim Berners-Lee FRS at CERN in 1989 (on which his supervisor wonderfully wrote “vague but exciting”).
It was Berners-Lee’s work – the mechanism that made it not just possible but also easy to share information – that kicked off the world wide web. His work was driven, however, by the need to share data with fellow scientists. Today’s increasingly data-heavy science leaves researchers with ever more complex versions of Berners-Lee’s problem. We have the infrastructure to produce petabytes of data. But often we lack the means to usefully share this information.
Our new policy study Science as a public enterprise, launched today, asks how we can improve access, for whoever wants it, to the evidence and methods of scientists. Open access to published research papers has paved the way, becoming a requirement for many research funders. But science goes well beyond its published papers and there is increasingly wide interest in much more than those artefacts. And with good reason: last week’s Times Higher Education featured the story of a scientific breakthrough that received interest from several top journals, but was subsequently derided because of major problems in the data and statistical analysis; and this article by David Dobbs illustrates how publishing can become a “choke point” rather than a platform for science.
It is not only in science that there is a drive to communicate more than a final, packaged product. Government transparency initiatives – President Obama’s data.gov and now data.gov.uk – hope to open up the statistics collected by Government departments. And a recent article in the Financial Times (SUBSCRIPTION) showed that the availbility of rich consumer data, rather than just sales figures, is creating a new industry in itself.
But for scientists there are clear challenges to the case for open data: how can they secure their authorship while sharing data? How much work do they need to put in to make data available? Whose responsibility is this? How can any personal data be properly annonymised? (See today’s Lancet comment piece by four of the study’s working group for a fuller discussion of these questions.)
When it is done right, however, there is clearly an appetite for opening up data. In a recent study of cancer microarray clinical trial publications, papers with publicly available data were associated with a 69% increase in citations. Tim Berners-Lee’s genius was to understand the importance of not just the existence of the internet, but the web of communication it facilitates. His motivation was that better science would be done when these connections are made. The new Royal Society study will explore what this web could look like for science in the 21st Century.
If you would like to make a submission to the call for evidence for this study, you can find more information here. Please tweet us your views @royalsociety using #openscience.
We are also holding a public town hall event at the Southbank centre on 8 June. Details are here.