Reinventing Discovery: The New Era of Networked Science

Michael Nielsen wrote Reinventing Discovery because he wants to light an almighty fire under the scientific community, and it's a good metaphor. Even after humans had discovered fire, for thousands of years life went on much the same. Then we invented science, the fire of the mind, and the speed of change increased dramatically. Modern science is now a collection of conventions about how communities of people think and solve problems together: it includes scientific papers, peer review and experimental methods, as well as concepts such as plagiarism and citation. A dramatic shift happened with modern science because of the business model that accompanied it: it is clear from scientific publications who does what and who gets priority for discoveries, and it is thus sound business to want to fund the science and spread the benefits around society. But thanks to the internet, we are reinventing how we make discoveries, and it is as radical a change as the discovery of the scientific method. It will speed up change many times over.

We had to wait until 1928 for Alexander Fleming to discover penicillin, and he did it by being lucky. If instead you used the internet to ask thousands of people to look for a useful drug, the process would be thousands of times faster. The probability of one person being lucky is small; with enough people, you can bet that one of them will be lucky. All you need to do is organise enough participants, as Nielsen emphasises here.

The Galaxy Zoo project, an inspiring example he considers, lets anybody classify galaxies. Instead of the classic lone astronomer painstakingly searching photographs hoping for a new discovery, thousands of people worldwide are looking at images online. One participant, Aida Berges, an ordinary 53-year-old living in Puerto Rico, has discovered two hypervelocity stars. It isn't luck; it's the result of 200,000 citizen astronomers working together.

We know the DNA sequences of millions of proteins, but we don't know how most of them fold up into shapes. Enter Foldit, an addictive online game that allows ordinary people to build and play with real proteins. Many work in groups, sharing insights. In scientific competitions, Foldit players are on a par with elite scientists and the best computer programs.

Wikipedia and Linux are more familiar examples of internet-driven community projects that have changed the way we work. Simon Singh's successful legal battle against the British Chiropractic Association is perhaps less well known. When the latter presented evidence that Singh had libelled it, bloggers helped demolish that evidence. Instead of arguments being presented only by conventional expert witnesses, the internet allowed many people to join in and scrutinise the issues behind Singh's case much more quickly and effectively.

These new approaches to doing science will transform the world, but we need cultural and business models to change, too. Open access is the key. In fields such as physics, it has revolutionised how quickly good science can be done. Free systems such as arXiv make the latest physics free to everyone, and everyone benefits. Sadly, in medicine and climate science, the science is mostly only available to those who can pay for access. This is cruelly ironic, as the new methods increasingly use citizen power to do the science - and in medicine, whole countries contribute human data, so why don't we all have free access to the results? Fortunately, research funders are starting to oblige scientists to make their work more freely available.

Reinventing Discovery will fire up scholars and scientists to make better use of the internet and join the open science movement. Nielsen neglects to cover topics such as combinatorial chemistry (which does much the same as many of the examples he cites, but doesn't need humans), but he'll inspire anyone to follow his pointers on the internet; it's all out there!

His real contribution, however, is his informed discussion of the social pressures slowing this process of reinvention. Unfortunately, there isn't space in a brief review to highlight all of them, from counterproductive incentives for scientists to hold back information from peers, to the shifts in scientific culture away from the idea of "the paper" as a researcher's passport to a good job. Nielsen offers keen insights into how legal, business and academic culture clashes with the pursuit of open science. Our pre-internet thinking is chasing short-term and narrow competitive benefits at the expense of the wider world.

What has been called "ubiquitous human computing", the splitting-up of tasks so that people around the world can contribute, will have as profound an impact on society as Henry Ford's production lines. It is a truism that the world's problems can only be solved by thinking. Now we know how to do it better.