We knew for quite some time that Michael Nielsen (blog, Twitter) was writing a book about the future of science. He said so on his blog a few times, and wrote some posts as early, preliminary thoughts on the topics he'd like to cover in it, back in 2008 and in 2009 (those two posts are especially worth re-visiting - still current and thought-provoking).

The book is now out, and I have read the galley proofs sent to me by the publisher early on (bloggers get lots of free books, with no strings attached - it is up to us to review positively, review negatively, or not review at all). I was just too busy to sit down and write the review until now...

For those of you who do not have the patience to read the whole review, but are interested in the way new technologies, especially the Internet, are changing the way science is done, I can say - go now and buy yourself a copy of Reinventing Discovery: The New Era of Networked Science. It is excellent. Worth your investment in time and money.

It is almost easier to describe this book by stating what it is not...

It is not an utopian, hyper-optimistic manifesto for all things Open Science (though the tone is generally optimistic and positive).

It is not an encyclopedia listing thousands of Open Science projects that people have started over the years, so your favourites (including perhaps a project of your own) are unlikely to be mentioned there.

What it is, is a careful examination of the potential for new ways of producing new knowledge. Note that it is not titled "Reinventing Science" or "Reinventing Research" but "Reinventing Discovery", a much broader concept than just science. It is about various ways in which groups of people, by finding each other online and working together (either cooperatively or competitively), can make new discoveries about the world.

What Michael Nielsen did was carefully choose just a few projects, each using different technology, each having a different goal, and each having a different method, and analyzed how and why they were successful. What small differences made these projects successful while many similar ones were failures? What aspects of the method were key to its success, what were the strengths and what were the weaknesses? Why are some methods better for some specific goals than other methods? Why were some - initially exciting and heavily touted - projects failures in the long run? What did they do wrong?

His most thoroughly described and analyzed examples include the Polymath Project, Galaxy Zoo, Linux, and the Kasparov vs. World chess match. Where needed, he also introduced other projects or classes of projects, e.g., Open Notebook Science, Open Access Publishing (e.g,.PLoS), Wikipedia, arXiv, GenBank, InnoCentive, Open Dinosaur Project, eBird and Fold-It. Some notable failed projects are also mentioned and analyzed (e.g., Qwiki).

As you can see, that is quite a breadth of projects with some very different histories, and very different goals. What ties them together is the crowdsourcing element - how large groups of people can solve problems that a handful of experts cannot solve alone, either due to lack of brain power, or lack of computing power, or lack of time. Some of these projects require that the crowd consists of people with high level of expertise on the subject - after all, non-mathematicians could not be of much help in the Polymath Project, or non-coders for developing Linux. But for other projects, literally anyone can contribute - one only needs to have a source of electricity, online access, and an instrument (e.g., computer, or smartphone) to get online to participate. The former are highly informal networks of experts working on a problem, the latter are what we now call "citizen science" projects.

What is notably missing from the book are lengthy discourses about science communication. There is nothing about the way the Web is disrupting and changing science journalism (and actually very little about scientific publishing). Science blogging is covered in a single subheading about one page long (though pros and cons of the blogging software for discovery are carefully dissected in the discussion of the Polymath Project). There is almost nothing about the use of the Web in science education.

And this is fine - covering all of those topics would make the book twice as long and would dilute its message (if I was not so busy and so ADHD, I'd relent to the hounding publisher/agent who wants me to write a book about those other areas - perhaps one day I'll say yes to that idea so Amazon.com can suggest that people who like my book should also buy Michael's and vice versa). The book is about the way knowledge is made, not how it is disseminated once it is made - there is an unstated undertone that the open online activity of the discovery, especially when it involves thousands of non-experts, will inevitably result in the spread of the information, coupled with the excitement of watching - and sometimes participating in - the way the information is wrestled from Nature.

The world Nielsen describes is quite reminiscent of the way science is done in "Rainbows End" by Vernor Vinge:

...How does one get answers to scientific questions, or get new technologies developed? By using the hive-mind. There are online boards and forums. You go there, offer virtual money, and the collective effort of the people on there provides you the answer in a timely fashion. It is so powerful that you can rely on the people to design you a new technology according to your specifications, and do it in time for you to go ahead with your plans, certain that the technology will be available to you at the time when you need it...

Just like in the fictional world of Vinge, the projected future world of Nielsen has in it a place for scientists trained in a traditional way, employed by traditional institutions, funded in traditional ways. After all, much of research is expensive, requires large investment in infrastructure and generally cannot be done at home in one's garage. But, like Vinge, Nielsen imagines a world in which those scientists are not isolated from the rest of the world - they are dependent on the broad participation of many other people, with varying degrees of expertise: some collect data, some are good thinkers, some lend their computers to it, others provide a little extra funding for particular small projects. This is a vision of a world in which science is just one integral part of the general process of discovery of knowledge, which is one integral part of what the world normally does every day anyway.

But unlike pure speculative fiction, Nielsen's ideas are built on a careful analysis of the past - from the anagrams of Galileo and Newton, to Henry Oldenburg and the invention of the scientific journal, to the invention of peer-review in mid-20th century, to the developments of the past couple of decades since the invention of the World Wide Web. It takes into account people and how they, being human, resist or accept new ways of doing old stuff. It points out the obstacles, and errors one can make in pushing for a more open and more collaborative research. But it also provides a blueprint for how to do it right. And this last thing is why YOU should buy this book and read it carefully - it gives you a cool-headed, calm, thoughtful analysis of the things that work. Use them.