Reinventing Discovery: The New Era of Networked Science

by Michael Nielsen

In Reinventing Discovery, Michael Nielsen argues that we are living at the dawn of the most dramatic change in science in more than 300 years. This change is being driven by powerful new cognitive tools, enabled by the internet, which are greatly accelerating scientific discovery. There are many books about how the internet is changing business or the workplace or government. But this is the first book about something much more fundamental: how the internet is transforming the nature of our collective intelligence and how we understand the world.

Reinventing Discovery tells the exciting story of an unprecedented new era of networked science. We learn, for example, how mathematicians in the Polymath Project are spontaneously coming together to collaborate online, tackling and rapidly demolishing previously unsolved problems. We learn how 250,000 amateur astronomers are working together in a project called Galaxy Zoo to understand the large-scale structure of the Universe, and how they are making astonishing discoveries, including an entirely new kind of galaxy. These efforts are just a small part of the larger story told in this book--the story of how scientists are using the internet to dramatically expand our problem-solving ability and increase our combined brainpower.

This is a book for anyone who wants to understand how the online world is revolutionizing scientific discovery today--and why the revolution is just beginning.

Genres: Science, Nonfiction, Technology, History, Internet, Computer Science, Computers

272 pages, Hardcover

First published January 1, 2011

Reviews

[Courtney Johnston · Rating 2 out of 5]

An important book, which lost its (tenuous from the start) grip on me on page 78.

Nielsen is an advocate for open science, and in this book he draws a picture of science standing at the threshold of its most important advance since the establishment of the Royal Society and the first norms of scientific publishing and data-sharing. The amplifying power of internet, he argues offers new opportunities for collaboration and sharing. The challenge is to move the bulk of the scientific community away from their closed and guarded approaches in order to take advantage of these opportunities.

Nielsen is clearly fired up:

These tools are cognitive tools, actively amplifying our collective intelligence, making us smarter and so better able to solve the toughest scientific problems. To understand why all this matters, think back to the seventeenth century and the early days of modern science, the time of great discoveries, such as Galileo's observation of the moons of Jupiter, and Newton's formulation of the laws of gravitation. The greatest legacy of Galileo, Newton, and their contemporaries wasn't those one-off breakthroughs. It was the method of scientific discovery itself, a way of understanding how nature works. At the beginning of the seventeenth century extraordinary genius was required to make even the tiniest of scientific advances. By developing the method of scientific discovery, early scientists ensured that by the end of the seventeenth century such scientific advances were run-of-the-mill, the likely outcome of any competent scientific investigation. What previously required genius became routine, and science exploded.

Such improvements to the way discoveries are made are more important than any single discovery. They extend the reach of the human mind into new realms of nature. Today, online tools offer us a fresh opportunity to improve the ways discoveries are made, an opportunity on a scale not seen since the early days of modern science. I believe that the process of science - how discoveries are made - will change more in the next twenty years than it has in the past 300 years.

The picture Nielsen draws is in many ways the opposite of that of Crick and Watson alternating between fervid bouts of creativity and languid cups of tea in the company of university popsies, stealing away with Franklin's x-rays and covering up their work as they rush for publication. He describes a number of projects that are models for the new kind of science he proposes, such as the Polymath Project (distributed mathematical problem-solving), open source software (the success of Linux, made possible through its modular nature, which allows a multitude of people to make innumerable small contributions), the Firefox bugtracker (which allows any user of this open source browser to identify issues and submit enhancements), and Kasparov vs The World (the grandmaster takes on the international chess community, who use online tools to suggest, evaluate and select their moves, drawing on their distributed specialised expertise).

I agree with Nielsen's argument, but I'm not inspired by his rhetoric. His book is largely pragmatic, which is really a rather wonderful thing - polemic will only get you so far, and I can see this as a book that one scientist might press upon another as the open side tries to win over the closed.

But as a pure reading experience, the book is like a rather stodgy and dull boiled pudding, studded with the odd tasty bit of crystallised fruit. Overal, it is delivered more like a lengthy lecture to an undergrad class than a book. The pages are peppered with phrases like these:

"Earlier in the book we discussed the open access policies that some of the scientific grant agencies are introducing...."
"We'll now look at two strategies tan can be used to shift the culture of science...."
"There are lots of ways of this is happening; so let me describe just a few snapshots..."
"I won't make all the connections explicitly, since this isn't a textbook on political economy. If you're interested in exploring the connections further, please see 'Selected Sources and Suggestions for Further Reading', beginning on page 217."

Those are all from a single page (admittedly, from the finl chapter, which I flipped to this morning just to see whether I should be slogging through the intervening pages rather than kicking it in). They're harmless in themselves, but they accrete in this puddingish texture, where over all the book feels earnest rather than inspiring. And that's not a note I want to end on, as I genuinely think the book is important, Nielsen is very very smart and persuasive, and the possible future he outlines is one that makes moral and intellectual sense. I wish I could have enjoyed the book more.

[Jovany Agathe · Rating 5 out of 5]

I read Nielsen's new book cover to cover on my flights to / from an Open Access Week event in Tucson this week and I give it my strongest recommendation for a pleasurable read about a crucial topic. I am a scientist and my students and I practice open science as much as possible--open notebook science, open protocols, open data, open proposals, etc. I have also seen the author, Michael Nielsen speak a couple times, and I have read many of his blog posts. So, before reading this book I didn't necessarily expect to learn much or certainly to be further convinced of the possibility of transforming science in this new era. From the moment I started reading, though, I was captivated. Many of the stories were not new to me (such as Galaxy Zoo or the polymath project), but I hadn't heard them in such detail before and I enjoyed learning a lot more about those successful crowd- or citizen-science projects. There were also many success and failure stories in open or collaborative science that I hadn't known about, such as the Microsoft-sponsored "Kasparov versus the world" chess event, or the research into how small groups can make bad decisions if the collaborative conditions aren't set up correctly. I learned a lot from these new stories, and remained captivated throughout.

[Amir-massoud · Rating 4 out of 5]

This book is about how we can/should do science from now on. It promotes Open Science approach, which is based on the ideas of sharing data in an open source fashion, using the network to focus the attention of experts, benefiting from intelligent amplification tools, and etc.

I think it is a must-read book for professional scientists and a good book for science enthusiasts. It is generally written very well. The downside is that at points it becomes repetitive and loses its fast pace. Even though the book is already very short (the main body is less than 200), it could still be compressed a bit more. In general, I am happy that I read it and I will probably go back to it in the future.

[Charlene · Rating 5 out of 5]

If you were one of the many people who excitedly picked up a copy of "wisdom of crowds" only to be disappointed when you realized that the passion with which the author wrote was matched only by the confirmation bias that accompanied it, then you will be extremely happy about this book. It's too looks at the role of collaboration in generating a finished product, but unlike wisdom of crowds, it is a solidly researched contribution to the field of network research.

The author looks at both the value and challenges of sharing data in the scientific community.

Great arguments, solid writing. I really enjoyed this book quite a bit and highly recommend it.

[Goran]

http://www.ted.com/talks/michael_niel...

[Amanda]

After reading the review in Science, I'd like to take a peek at this book myself. http://www.sciencemag.org/content/336...

[Liam · Rating 5 out of 5]

"In the most successful online collaborations this use of microexpertise approaches an ideal in which collaboration routinely locates ... people with just the right microexpertise for the occasion. In particular, as creative collaboration is scaled up, problems can be exposed to people with a greater and greater range of expertise... Instead of being an occasional fortuitous coincidence, serendipity becomes commonplace. The collaboration achieves a kind of designed serendipity..." (27)

"In this chapter we'll identify four powerful patterns that open source collaborations have used to scale. (1) a relentless commitment to working in a modular way, finding clever ways of splitting up the overall task into smaller subtasks; (2) encouraging small contributions, to reduce barriers to entry; (3) allowing easy reuse of earlier work by other people; and (4) using signaling mechanisms such as scores to help people decide where to direct their attention." (48-9)

"This points the way to a fundamental requirement that must be met if we're to amplify collective intelligence: participants must share a body of knowledge and techniques. It's that body of knowledge and techniques that they use to collaborate. When this shared body exists, we'll call it shared praxis..." (75)

"Citizen science can be a powerful way both to collect and also to analyze enormous data sets." (151)

"The problem today is that while it's now in the collective interest of scientists to adopt new technologies, their individual interest remain aligned with journal publication." (189)

[Matthewmartinmurray murray · Rating 4 out of 5]

Lots of fun to read. I started off really enthusiastic about this. Then it got to be a little bit repetitive by repeating its themes and same 3-4 examples too many times. Its still very interesting to learn about collective intelligence. The thesis of the book is to encourage science to take on a new paradigm of open source data collection and shared results. Seems to be a little idealistic but it makes a pretty good point of how fast innovation could advance if everybody is directed in a similar path and gives free time and energy to put into a universal goal. I believe this would work only if it is formatted in an interesting way that would capture the imagination of the volunteers and there would need to be some un-named incentive that hasn't been presented yet. It is fun to think of a world that can advance as quickly as a potential collective of scientist working for one goal. Its powerful what a group of people can do when properly motivated and directed. If any of this is interesting to you, then you may like this book.

[Piotr · Rating 3 out of 5]

Michael Nielsen presents how Internet enhances our knowledge and problem-solving skills. He provides examples how (and when) a well-done collective intelligence can outsmart the most capable minds and be used in the leading edge of science and technology.

Moreover, the book not only contains examples, but has a visionary part - it shows an already started path to science more open among scientists and also inviting amateurs for their meaningful contribution. If you are interested in the future of knowledge making and sharing (or at least, if you consider Wikipedia a great thing) - read it.

When it comes to drawbacks: the book seems to be written in a chatty style (call it 'light' if you like it), with some repetitions, unpolished parts and (IMHO) could be losslessly compressed to 1/2 of its volume. Moreover, it is kind of ironic that a book on open knowledge is only accessible as a copyrighted material. While I like having a paper book in my hands I am astonished there is not a pdf/wiki on the author's blog.

[Jani-Petri · Rating 3 out of 5]

Decent discussion about the open science and how to approach it given the constraints, for example, faced by the career demands of the scientists. In times I had the feeling that the writing was a bit sloppy and repetitive, but this was easy enough book to read and a useful way to start discussion for real.

[Akrabar · Rating 5 out of 5]

a masterpiece from a master

[Stephan Rasp · Rating 4 out of 5]

Nielsen’s book echoes my frustration with academia. He argues that the current state of academic science is untenable (it is!). An absolute must-read for any scientist.

[Matthew Feickert · Rating 4 out of 5]

Reinventing Discovery feels like a clear manifesto for my ideas on open science. For context, I am a physicist currently working on the ATLAS experiment at CERN and as a software researcher for the Institute for Research and Innovation in Software for High Energy Physics (IRIS-HEP), and I try to make as much of my work open by design. I found it interesting how many of the same points that Michael Nielsen brought up in the book (published in 2011) on the problems that plague science are still quite relevant today. I am biased towards Nielsen's writings though as we share common goals for open science and Nielsen's book on machine learning, Neural Networks and Deep Learning , was one of my first formal introductions to the subject and the arguments and presentation of machine learning "made sense" to me. Conditioned on that, and my my work angled at producing more open and reproducible science it is perhaps not difficult to understand why this section in the final few pages of the book in Chapter 9 resonated with me strongly:

If you're a scientist who is also a programmer, you have a special role to play, an opportunity to build the new tools that redefine how science is done. Be bold in experimenting with new ideas: this is the golden age of scientific software. But also be bold in asserting the value of your work. Today, your work is likely to be undervalued by old-fashioned colleagues, not because of malice, but because of a lack of understanding. Explain to other scientists how they should cite your work. Work in cahoots with your scientist programmer friends to establish shared norms for citation, and for sharing of code. And then work together to gradually ratchet up the pressure on other scientists to follow those norms. Don't just promote your own work, but also insist more broadly on the value of code as a scientific contribution in its own right, every bit as valuable as more traditional forms.

If I am to critique the book, I would have enjoyed some additional examples beyond The Polymath Project, Galaxy Zoo, and Kasparov vs. "the World", which are used extensively throughout the book to hammer home the arguments. However, this is perhaps unfair as I can believe that these examples were carefully selected to both make points and to use as touchstones to tie together the narrative. Also, I am not a historian of science — even recent science — and as things like GitHub were still growing and Zenodo didn't even exist at the time of publication (Zenodo was launched in 2013 according to Wikipedia) there might not have been as many examples as there are today.

Overall, I would say that if you're someone who has a passing interest in open science this is a great book to read. It also does a great job of shining a light on the real institutional and sociological problems that face science today in terms of moving forward in its goals.

[Lisa Kucharski · Rating 5 out of 5]

Really enjoyed this, it covers the groundwork of science being practiced in a different manner than in the past... some examples are the FoldIt and Zooniverse efforts where data is analyzed by citizen scientists and data used by scientists... but also engages non-scientists to discuss the work as well.

He covers some great ground where open data, involvement of the public, and a forum that encourages use and discussion amongst both highly studied scientists and laymen. Of course there are only a few of these projects primarily as - scientists aren’t rewarded to create these kinds of projects in general. So, the few that are there are unique in their area. Generally the data is shared via the internet and anyone can come and help, search and just peruse the information.

Nielsen also goes over the pitfalls of the use of information, the reality of how scientists work and of the influences to not share currently out weigh sharing.

But what is most interesting is that he talks about how this could be a new chapter in doing science... where data is openly shared, through a forum that allows discussion, helpful mentors and a focus on developing questions to discover new information.

He does seem to indicate that this type of work does better with sciences that don’t involve “social sciences” since opinions of what is right or wrong tend to get in the way of clearly looking at data- without pre-conceived answers.

I find the ideas presented really interesting and could be a great way for humanity to solve some bigger problems that effect us all... but we collectively- humanity plus scientists must decide that we all need to move this way at the same time- in ways that benefit us all. Could be a great way to solve disease, but the money made by preventing knowledge of cures would certainly squash that idea.... sad but true.

Worth reading.

[Esben Kranc · Rating 5 out of 5]

Michael Nielsen has some absolutely fascinating views on human scientific development and the ideas from Reinventing Discovery are completely in line with what I'd love to see modern science become; a networked, well incentivized, decentralized system for knowledge generation that collaborates and engages the world. Let public be what public is, and let government-funded research be open to all.

With a close look into specific open tools, citizen science, democratization of research and data, open science projects, and blogs, Nielsen ends the book with the Open Science Imperative, calling upon researchers, universities, and specifically funders, to create wide agreement and decisive change on researcher incentives. It is a massive change from publication-based knowledge generation but he compares it to the lane change in Sweden 1967, where the whole of Sweden began driving on the right lane instead of the left. A drastic change *can* happen.

For Sweden, it took 10 years of discussion to reach their shift, while I see that since this book was released, many knowledge-generating processes are changing: OpenReview, ArXiv, codebase citations, gold standard open source, many research bloggers, a separate social incentive system, independently funded researchers, and so much else.

[Keith Elliston · Rating 2 out of 5]

I read this book expecting a thoughtful discussion of how we can use crowdsourcing and networking to improve life science. What I read was a very poorly edited book (so many typos that I stopped counting) that tries to apply analogies from mathematics to biological science that simply do not work. As many have said in science, ideas are a dime a dozen, its really a matter of doing the work to produce new science. The challenge in the empirical world of science are not so much the generation of new ideas, but the development and testing of hypotheses. The author is clearly well versed in the fields of math and physics, but fails to understand the challenges to doing empirical science in biology. The book could have used a thorough editing (I expected better from Princeton University Press) and the insights of scientists from biology and chemistry. I would love to see a rewrite of this book after having the author explore how empirical science is done... this world has been networking in various ways for hundreds of years. Technology is enabling new ways to do this, but they are simply not explored in this book.

[Emil Petersen · Rating 3 out of 5]

This is a nice introduction to the way relative recent technologies (basically the internet) afford a better way to collaborate and do science. Roughly, in today's research environment, the rewarded behavior is to publish frequently. The problem with this is, among other things, that time spent developing collective tools or time spent making discoveries and the associated work/data available, does not really advance personal publication. Think of it in terms of game theory: the people who selfishly devout all their time writing papers and doing research will do better in today's reward system than people who spend time sharing their results and tools, etc. This is a problem, because the community as a whole is better off if people share results and help each other. Michael Nielsen offers some potential solutions and lays out the problem much better than I just did; it is definitely worth a read if you're interested. The book is a few years old and my impression is that the problem is less prominent in the field of computer science, which is where I spend most of my time. Still, the problem very much persists today.

[Atila Iamarino · Rating 4 out of 5]

Me recomendaram muito este livro por conta de uma apresentação sobre ciência cidadã que fiz (ciência feita por não cientistas). Li e gostei bastante, mas ele falou exatamente o que eu já estava apresentando, não me acrescentou muito.

Michael Nielsen discute como colaboramos muito mais hoje em dia graças a internet e o tipo de produção científica que é possibilitada. A descrição do Galaxy Zoo e de como as pessoas estão fazendo descobertas astronômicas de casa, classificando as imagens, foi especialmente marcante.

Vale para cientistas se interessarem por acesso aberto e pela importância de comunicar melhor o que fazemos. Mas ele bate no ponto mais forte, enquanto o sistema de incentivos não mudar e não passarmos a valorizar esse tipo de atitude, nada vai mudar.

Ah, a descrição de Galileu contando o que havia descoberto por códigos foi especialmente marcante. A explicação fica para quem ler ;)

[Antony Mayfield · Rating 4 out of 5]

The book has an incredibly strong start – I was enthralled by its description of attention architecture, especially. The second third has a lot of detail about some interesting open science projects and then it moved to something of a polemic. An update would be very welcome as it is a few years old now and countless things have moved on in this field. If it were updated, I think the sections of the conclusion encouraging scientists to blog, share data etc might usefully be expanded to a section of protocol advice.

I’d like to know more about the ideas around attention architecture and collaboration. Four stars for the compelling and inspiring exploration of these. Highly recommended short read for non-scientists and scientists alike who are interested in how we can evolve new ways of working online.

[Gponym · Rating 4 out of 5]

From browsing the clear and lively Reinventing Discovery you might not guess that Michael Nielsen is a physicist and well-known contributor to quantum computation theory. However, if you delve into the book, you'll soon detect the careful thinking patterns of a trained analyst or scientist. Nielsen put his scientific career on the back burner in 2008 to focus on bringing about a revolution in how science is carried out (https://en.wikipedia.org/wiki/Michael...). This book is one fruit of his new mission. It aims to sway scientists as well as to educate the public and appeals both to logic and emotion. Nielsen is pushing all of us toward what he hopes will be a leap in scientific progress built on an explosion of online collaboration and open data sharing among scientists and even the interested lay public.

The title expresses Nielsen's belief that we are poised on the cutting edge of a fundamental reshaping of how most science gets done. Using some fascinating case studies he illustrates how this new era has already crept onto the scene. He cites the Polymath Project as successfully demonstrating the power of online collaboration. In this case an important mathematical challenge was solved entirely in the public eye over the course of a few weeks in 2009. Over 20 mathematicians contributed jointly to a web blog devoted to the problem that was hosted by Tim Gowers, a highly renowned mathematician. It was a heady time: Gowers later said that the Polymath process was "to normal research as driving is to pushing a car" (p 2).

Nielsen reviews the legacy process of scientific discovery which, he notes, has worked well yet whose essence has changed little since the advent of the first scientific journal in 1665 in England. He goes on to show the power of new resources and mechanisms that can support shared discovery (as in Polymath) but whose adoption by the broad scientific community lags. He asserts that the solution of many important scientific questions now and in future would benefit by the deliberate, thoughtful harnessing of the collective mind, and by openly sharing data, both research results as well as tips and informal knowledge about how to organize and carry out research. Nielsen foresees the need for creative and relentless design and redesign of collaborative platforms, for careful structuring of data and query mechanisms to facilitate automated access, and for open sharing of scientific data on a scale that dwarfs even current efforts like the Human Genome project or the Sloan Digital Sky Survey (SDSS) (p 98).

Nielsen lucidly presents many samples of these new ways of doing science, from Polymath to the Hap(lotype) Map to Galaxy Zoo (a website where anyone can help classify celestial objects) to the astounding Kasparov versus the World chess match in 1999. He draws on appropriate arenas outside science, including the software Open Source movement, to discuss advanced practices for enabling rapid and efficient online collaboration. "Architecture of attention" (p 32) and "designed serendipity" (p 27) are two memorable phrases he employs to tag the essence of these new practices. He contrasts all too briefly the new to some established ways of collaborating: conventional (hierarchical) organizations; the marketplace; offline small groups. He launches the effort to distil some principles that could guide pioneers who wish to implement online collaboration in a new problem-solving arena; it's but a tantalizing glimpse into what must become a well-explored domain if the new collaboration style is to catch fire. He sketches a dream of one day having an open data web that can machines can query on their own. He notes that with such a large trove of accessible data, our notion of "explanation" may need to change: no longer will Occam's Razor be as sure a guide to what is a good explanation, since machines can develop explanations that no human can. Without digging deeply into it, he notes that inevitably new types of questions can be asked and answered when large troves of machine-accessible data begin to jostle shoulders in the public domain.

The book, then, is a survey of the landscape and some possible ways forward. It feels like just an opening manifesto in a campaign that is in tune with large societal shifts such as social networking and putting online just about everything. It is supported by sections of notes, references, suggestions for further reading and an index. As noted, it is studded with compelling real-world examples. Scientific culture change is no minor matter and Nielsen deals with some obstacles. More inspiration than howto, despite the many examples, it does not address details on how to implement solutions in particular disciplines. Also, I suspect that much more fine-grained attention than Nielsen has given will be needed to pick apart and tackle the significant personal and money-related resistances that hamper adoption of his proposals. The large amount of for-profit science is hardly mentioned here; as an outside observer, I wonder if that is so trivial an omission in this age dominated by large, rich corporations.

No book can do it all; Reinventing Discovery makes a good initial exploration of possibilities to enhance the process of discovery. This book should appeal to anyone interested in how science "happens" (or doesn't happen!), in how to potentially accelerate progress without huge financial investments, in novel ways to tackle many classes of problems.

[David · Rating 3 out of 5]

Mostly focused on Open Science and Citizen Science, gave me better idea of what the vision of Open Science looks like and what was its history and perhaps also what to prioritise in open science (e.g. open data much more important then open access to papers; need to acknowledging alternative scientific outputs aside from papers as outputs worth recongnition).

Interesting but not that full of new ideas and concepts as I hoped, also the title is a bit misleading I think

[Ana · Rating 5 out of 5]

This is a great book - he talks about science and gives many excellent cases for the open science initiative.
However, sometimes the author details too much about some things - he uses a whole page for some things that could be said in 2 sentences.
Anyway, would recommend!

[Vittorio · Rating 5 out of 5]

everyone should read this book, which is not a book, it is an injection of power that opens your mind. I am not joking, if the meaning of this reading was grasped and applied in all sciences and in the social, in politics and in the economy, in school and in medicine, we had a better world

[Quentin · Rating 5 out of 5]

The most important ideas are covered early on but this is a short book and the anecdotes and arguments that follow are as entertaining and inspiring as they are thought-provoking. Possibly one of the most annotated books I own and surely one that defined my path and thinking going forward.

[Nikolay Theosom · Rating 3 out of 5]

quite re-iterative and doesn't really offer anything beyond repeating mass media news

[Adam Catto · Rating 5 out of 5]

Simply wonderful. If you care at all about the progress of science, technology, or humanity, then read this book ASAP.

[Rohit Ramakrishnan · Rating 5 out of 5]

Michael Nielsen was one of the authors of the book which first introduced me to the world of my passion- Quantum Computing. It was disheartening to learn later that he quit the field. But reading this book makes me immensely happy and provides the necessary insights for a scientist, though he talks nothing much about science in this one.

In the era of “publish or perish” and when research becomes more and more commercialised or secretive, Nielsen calls for openness in scientific research from the perspective of public. The book is the need of the hour, especially at the dawn of quantum technology which could profoundly change human civilization. Highly recommended for everyone who interested in science research.

[Lily · Rating 3 out of 5]

This book campaigns for open science: a scientific culture in which data and ideas are shared more freely, with the goal of facilitating large-scale collaborations among people with diverse scientific backgrounds (including people without formal scientific training). A key idea is that oftentimes a large scientific task can be divided into smaller, modular, specialized tasks. These tasks can then be taken on by experts in specific areas, thus joining the knowledge of many individuals into a larger collective intelligence that can move science forward in creative new directions that would be otherwise inaccessible. Much attention is given to the tools that enable the interactions needed for open science - wikis, databases, forums, and even games. One of the strengths of the book is the abundance of concrete examples that show where open science has succeeded, as well as where it has largely failed (thus far). These examples are informative for anyone interested in applying the strategies discussed in this book to their own work. I also appreciated that the author briefly touched on how scientific explanations are no longer "constrained by the limitations of our own minds": computational techniques have made it possible to develop complex explanations in the form of models, and in many cases these models have greater predictive power than simple intuitive explanations expressed only as words or equations. It's an exciting prospect, but one that hasn't been fully appreciated in some fields.

Despite these interesting points, I was unable to enjoy this book because of how clunky the text is. It's verbose and repetitive. I think it could easily be edited down to ~1/2 of its current length without any loss of real information. This issue is related to the book's overly conversational tone. There are several digressions that are presented in a rather self-indulgent way, usually preceded by (more or less) "I can't resist sharing this tidbit...". All too often (I lost count of how many times), the author instructs the reader to "imagine" some scenario that is at best tangential (and at worst irrelevant) to the point being made. Lastly, certain topics were given an oversimplified treatment, such as the acceptability of a "preprint" as a unit of publication - much more true of physics and computer science than biology and medicine. I would have liked to see a more serious discussion of some of the potential pitfalls of open science approaches - as it is, the author largely brushes them off while reiterating his optimism. In many cases it felt as though the author was just instructing the reader to adopt his point of view, rather than to consider his arguments and develop their own opinions. All of this made for a somewhat condescending (although earnest) tone.

Even so, I do give the author credit for raising important issues and for challenging the purely paper-centric system that scientists currently exist in. Arguably, we could be in a better state if more appreciation was given to people who develop and share tools/data, if scientific research was made accessible to more of the general public, and if people from different disciplines would pool their intelligence more often. For these reasons, I'd give the book three stars rather than two.

[Joshua]

An enjoyable read. Nielsen looks at ways of doing science in new ways leveraging groups and openness. The keys to success that he outlines;

- Tapping micro expertise (e.g. Innocentive)
- Modularizing collaborations
- encourage small contributions
- build a rich community of past data and debate (i.e., design for "reuse") (e.g., Polymath)
- dynamical division of labor
- use signaling cues (e.g., scores) to direct attention. This works best if it can be computed immediately.
- users have to have a shared praxis (knowledge and techniques) for this to work (e.g., wouldn't work for literary theory)
- there is a peril of partial information with low-status workers deferring to high-status workers. Avoid this! (e.g., pseudonyms)
- be sure to indicate speculative comments and a mechanism for acknowledging mistakes
- develop tasks that are challenging, but not impossible, provide instant feedback on how well you're doing, and seem like you're always just a step away from improvement (e.g., FoldIt)
- You have to take a leap to do this, it is not necessarily encouraged institutionally (yet). Funding agencies can help by making data sharing obligatory and rewarding largescale colaborative work/outreach. So can peer-pressure.

One thing I wonder about is whether you can use this same framework in teaching a course. The idea would be to turn the "classroom" into a sapce for discovery. E.g., creating a polymath-like structure to organize a junior-level class (e.g., PChem) around solving a problem. the goal would be to develop the shared praxis, but in the context of solving problems that would tap the emerging microexpertise of the group. Haven't really thought this out completely, and it sounds really tough to do well...but would be great.

[Pete · Rating 4 out of 5]

Reinventing Discovery (2011) by Michael Nielsen is a book that looks at how the Internet and advances in computing technology is changing and will change the way that science is done.

Nielsen starts by looking at the Polymath Project, a blog set up by Tim Gowers, a Fields Medalist, where math problems have been solved at amazing speed. He goes on to mention the Galaxy Zoo and the Sloan Digital Sky Survey where amateurs can do valuable work in identifying galaxies and can do it with remarkable speed and quality. Nielsen makes the point that similar to the way in which Open Source has transformed software Open Science can improve science.

The book looks at various examples of how the Internet has enabled people to collaborate and create in ways that were not previously possible. ASSET, the open source movement and Linux and Wikipedia have all been highly successful.

Nielsen looks at how the publish or perish mentality of science works against open collaboration and how this can be overcome. He also looks at the rise of free access publications like ArXiv and PLoS have arisen and how they are changing academic publishing.

The book unwisely looks at Climate Science and how according to Nielsen this shows the limits of open access to science. He states that the quotes from the Climategate are all being taken out of context. He would have done far better to either avoid the subject entirely or to look at the way the blog Climate Audit studied climate reconstructions in the open and the transparent way in which climate model output has generally been made available.

The book is thought provoking, well written and interesting. Nielsen goes over his themes carefully and with deep knowledge of them. It’s fun to read and informative.