Spreading Science Knowledge Far and Wide

Seven Science 2.0 pioneers offer their perspectives on science, information sharing, collaboration, and the role of technology going forward.

Published May 1, 2010

By Adrienne J. Burke
Academy Contributor

Surely you’ve noticed: The scientific community is undergoing a research-and-data-sharing sea change. Perhaps slower to take to Web-based dissemination than some professions, science—the endeavor for which the World Wide Web was developed—has gradually been adopting new online methods for distributing knowledge. Some say the changes could accelerate scientific progress.

From open-access journals to research-review blogs, from collaboration by wiki to epidemiology by Blackberry, networked knowledge has made more science more accessible more quickly and to more people around the globe than could have been imagined 20 years ago.

And it’s not just new media businesses that are pioneering the Science 2.0 movement. Traditional scientific journals are part of this social evolution too, innovating ways to engage scientists online and enable global collaboration and conversation. Even the 187-year-old Annals of the New York Academy of Sciences has joined the digital age. The Academy now permits free public access to selected online content and has digitized every volume dating back to 1823.

That wider, freer, faster access to scientific data and research results will benefit the world is, to many, intuitively obvious. “We work on the assumption that the reason we publish is to keep science moving forward,” says Public Library of Science founder Harold Varmus. “If everybody can see the work that we do, and new work is built on what’s come before, science moves faster.”

Varmus is among a cadre of iconoclasts calling for immediate open access to scientific papers. They’re impatient for colleagues to give up their allegiance to the conventional process that they say keeps new research under wraps for too long. And they’re eager for publishers to break out of business models that require a paid subscription to read the most current publications.

To be sure, some changes are easier advocated than adopted. The most esteemed peer-review journals have taken great leaps toward openness in the last decade. Some now help readers network with each other online or enable posting on their Web sites of commentary and conversations about scientific publications. Many make papers openly accessible after a certain time. But how to sustain a business that publishes peer-vetted, high-quality content without requiring payment for access remains a hotly debated question.

As Varmus himself points out, the essential importance of the scientific paper has a lot to do with why it’s not just for-profit publishers, but scientists themselves who are moving toward open access with such caution. “Publication is not an addendum to, but the heart of the career of scientists,” he says. “The way you’ve built a legacy is through your publication—it’s the most important thing you do.” To give up the emotional reward of seeing their research published in a distinguished journal is a lot to ask of scientists raised in this tradition.

Seed Media Group CEO Adam Bly hints at how the up-and-coming generation of scientists—the so-called “digital natives” who’ve never known a world without the Internet—might move science past the paid-access paper. Says Bly, “In a Seed research study, one scientist said to us, ‘The soul of your identity is on the Web, because it is your most direct form of communication out into the wide world. You have a great degree of control over how you present yourself, your ideas, and your findings, and it’s fast, and it’s free.’”

For help considering whether the desire for open access contradicts the value of peer evaluation and envisioning what the future of science publishing could look like, The New York Academy of Sciences spoke with Varmus, Bly, and five other pioneers at the forefront of the Science 2.0 movement. These experts in Web technology, publishing, law, and science have the vision and passion to change the future of the way you work. As Bly says, “Open science is not this maverick idea; it’s becoming reality.”

Harold Varmus: Co-founder and Chairman of the Board, Public Library of Science

Harold Varmus, a Nobel Laureate, President and CEO of Memorial Sloan-Kettering Cancer Center, and member of the Academy’s President’s Council, led the team of biomedical scientists who set out in October 2000 to liberate access to scientific research in their field by petitioning publishers to post peer-reviewed papers in free, public online archives.

Varmus and his cohorts ultimately launched a nonprofit open-access publishing venture, which achieved financial sustainability this year. The Public Library of Science journals—there are now seven of them at www.plos.org—make scientific papers immediately available online, with no charges for access and no restrictions on subsequent redistribution or use, as long as the authors and source are cited, as specified by the Creative Commons Attribution License.

Would you define what you mean by “open access.”

Some people think that if their content is online it’s “open access.” That’s not the case. “Public access” is what the National Institutes of Health now operates under; if your work is supported by the NIH, then you must be sure that it’s available in less than a year on a public database like PubMed Central. That was a big victory for us, but it’s not anywhere near the goal.

True “open access” is different from “public access.” It means that the author holds the copyrights, that the journal places the work immediately and freely in the public domain under a Creative Common license or something equivalent to it, and that the work is in public libraries and available for all kinds of reasonable use, as long as attribution is maintained.

Have scientists been slow to embrace submitting their work to open access journals?

There’s now pretty wide acceptance of Public Library of Science journals, but most of my colleagues are still tormented by the need to publish in Nature, Cell, and Science, which are not open access journals. This is about much more than just publishing; it’s about values in the scientific academic community. Biomedical trainees are completely obsessed with the idea that they can’t get a job unless they publish papers in Nature, Cell, and Science. This is unfortunate, because those journals are going to be the last to go completely open access.

PLoS is now publishing far more research than any of those journals, isn’t it?

Yes. We publish over 600 articles a month. The only way you really can change the culture is to take on those top journals, so we decided we would publish two journals, PLoS Medicine and PLoS Biology, to compete with the very best.

We’ve achieved a high level of credibility for PLoS Medicine and PLoS Biology. They’re so-called high-impact journals. But to do that means rejecting a lot of articles, which gets expensive because of the costs of reviewing articles that do not get published. We afford those two journals because we make very modest amounts of money from other higher volume journals and we cover the cost of the whole enterprise by balancing things out.

What about the importance of the impact factor in scientific publishing?

The impact factor is a completely flawed metric and it’s a source of a lot of unhappiness in the scientific community. Evaluating someone’s scientific productivity by looking at the number of papers they published in journals with impact factors over a certain level is poisonous to the system. A couple of folks are acting as gatekeepers to the distribution of information, and this is a very bad system. It really slows progress by keeping ideas and experiments out of the public domain until reviewers have been satisfied and authors are allowed to get their paper into the journal that they feel will advance their career.

What are some ways PLoS is taking knowledge-sharing to the next level?

One of the most important developments is not particular to open access journals, and that is the addition of online commentary. Here’s our opportunity to make every article an occasion for conversation and a way to have another kind of evaluation. I can imagine search and promotion committees of the future spending more time looking at the kind of commentary that a paper has elicited than calculating impact-factor scores.

We’ve tried another experiment in the last few months called PLoS Currents. We’ve done this with one subject so far—influenza, a topic of great interest with a need for rapid publication. We invite people to post in PLoS Currents anything that can be looked at by a board of curators in 24 hours. The point is to get an article or an idea or a single result into the public domain quickly so people can build on it.

Look at PLoS Currents: Influenza on our Web site and you’ll see it’s been quite a nice experiment. Some postings look like full-fledged articles, others look much more primitive, but most have anywhere from a few to 10 or 20 commentaries attached to them. This is a way for scientists to get others to comment while they’re still working.

Information can also be aggregated and put together in very useful ways on sites that we’ve been calling Hubs, a project still in development. The idea is to try to wrest deeper ideas out of aggregated material without violation of copyright. We hope to create communities that migrate to these sites every day and then use them as platforms for fostering their field. This is another way to make science more energized.

What would be one technical fix you’d wish for right now to enable more sharing of science?

We have problems about sharing in our community that are not very technical, and it’s important to keep those in mind. Getting people to share their reagents, their mice, their plasmids—there’s a problem. People seem to forget that they were paid by the government or by some charitable agency or an institution to do this work and that they don’t own it. Say you made a new transgenic mouse 10 years ago or even two years ago and somebody else wants it.

You ought to give it to them, and you don’t need cloud computing to do that. Before we make all sharing digital, let’s remember that there are some simple things that reflect community values that we don’t subscribe to with the kind of enthusiasm we should. Of course, we’d also like to see everyone publishing more papers in open-access journals, especially at PLoS!

Adam Bly: Founder & CEO, Seed Media Group

After a three-year stint researching cancer at Canada’s National Research Council while still a teenager, Adam Bly set out to launch a magazine to cover “the 21st century scientific renaissance.” Five years later, his Seed Media Group has expanded beyond its glossy print flagship, Seed, to launch several online products serving science, including: ScienceBlogs.com, a social media site reaching more than 2.5 million readers; ResearchBlogging.org, which aggregates and feeds to relevant journals blog conversations about the peer-reviewed research that they publish; and ScienceWide, a platform that aims to drive advertising dollars to support open-access science publications and other innovative online science tools. Bly’s company’s mantra: “We are inspired by the potential of science to improve the state of the world, and we make media and technology to help realize that potential.” 

What do you mean when you say that science publishing needs to adopt a digital core?

Science has gone digital. Open science is not this maverick idea; it’s becoming reality. About 35 percent of scientists are using things like blogs to consume and produce content. There is an explosion of online tools and platforms available to scientists, ranging from Web 2.0 tools modified or created for the scientific world to Web sites that are doing amazing things with video, lab notebooks, and social networking.

There are thousands of scientific software programs freely available online and tens of millions of science, technology, and math journal articles online. What’s missing is the vision and infrastructure to bring together all of the various changes and new players across this Science 2.0 landscape so that it’s simple, scalable, and sustainable—so that it makes research better.

How will that happen?

To affect this kind of change is a grand challenge and will take the participation of many stakeholders—from government agencies to funding bodies to scientists themselves. The next generation of PIs is already establishing new behaviors. They feel comfortable blogging, using social media tools, and using wikis to advance their research. It will take the big institutions to support open-access journals, for example. And it will take technological innovation in the form of software that is purpose-built for this unique community and its set of challenges.

The culture of science resists change to science itself, and it’s important that it does. Part of that is practical: nobody sets rules for all of science. So it might take 10 or 20 years or more to effect a complete transformation. We’re talking about something as fundamental and important as modernizing the architecture of science.

What are some ways your company is contributing to this transformation?

We’re listening to scientists and introducing software and digital and social media platforms to help spur and support this transformation. Any scientist who blogs anywhere can now go onto ResearchBlogging.org and download free software that we’ve built that allows them to easily affix to a post the digital object identifier (DOI) of the scientific paper they’re blogging about along with some metadata.

We’re aggregating all of the conversations that are happening around that specific paper, and, through ResearchBlogging Connect, feeding the conversations back to scientists and journals in the form of widgets and RSS feeds. Now, when you’re reading the paper online, you see a feed of blog posts associated with that paper coming from across the Web. So in this example, we’re tackling post-publication peer-review and working to connect analog to digital in a way that’s seamless and useful to the scientist.

It sounds like you could have a new way of measuring a paper’s impact.

There are a lot of people trying to bring forth some new ideas about how to create more dynamic indicators. There are people merging scientometrics with data visualization, and there’s amazing work being done at universities around the world to develop new ways of measuring scientific progress. One thing we’re really interested in at Seed is whether blogs and the conversations we’re now organizing can serve in any way as an indicator of the momentum of scientific ideas. Technology can afford us more dynamic intelligence and useful knowledge.

James Boyle: Founding Member, Board of Directors, Creative Commons

James Boyle is a widely published leader of the global discussion about the ways that current copyright, patent, and trademark laws stand in the way of innovation by interfering with access to information that is in the public domain. He was one of the original board members of Creative Commons, which works to facilitate the free availability of art, scholarship, and cultural materials by developing licenses that individuals and institutions can attach to their work.

And he was a co-founder of Science Commons, which aims to expand the Creative Commons mission into the realm of scientific and technical data. In 2000 he joined the faculty at Duke University, where he is William Neal Reynolds Professor of Law and co-founder of the Center for the Study of the Public Domain. He is also a board member of the Public Library of Science.

What do you see as the current problem with access to science knowledge?

Science knowledge generation has gone digital, but our method of knowledge processing is still analog. Most scientific literature is behind pay walls. You may be able to find it with Google, but you probably can’t read it. That’s Science 1.0: You don’t have access unless you’re sitting in a great research university where it’s free, and you certainly can’t send a robot to crawl the literature to create a mini index of all the articles, and cross index them and see whether, for example, a particular gene known by multiple names is referenced by them.

Is the prestige attached to publishing with closed journals part of the problem?

Right now, if your article gets into Nature or Science it’s a big help in getting tenure and grants and retaining grad students. That’s important—we should encourage people to publish. But perhaps we could refine the incentives so that you get more of a bump for publishing openly. I would like to see people’s resumes say when their database has been downloaded more than 1,000 times. You want the prestige economy to reward the prosocial behavior, not the anti-social behavior.

So, how can incentives be changed?

When you’ve got centrally funded science, it should be a pretty easy cascade to start. The funders get much more bang for their buck if they do this. You’re actually saving the public money and increasing the yield of every research dollar. 

Once the idea can be explained to people, it makes an enormous amount of sense. I tell scientists, “There are a billion people connected to the Web. At least one of them has a smarter idea about what to do with your data than you do.”

Their first take, though, is “Oh, great. You’re going to force me to annotate my data and put everything out there. You’re going to troll it and publish ahead of me. I’m going to get no credit, I’m not going to get tenure, and I’m going to end up living under a dumpster. And you’re going to win the Nobel Prize.” That mindset is the big obstacle.   

We need funders to say that a condition for the funding is data deposit in an open, accessible format. That’s beginning to happen—the public-access mandate from NIH is beginning to make the literature openly available. But we’re just at the beginning.

Beyond social/cultural issues, what else needs to change?

Nobody ever wants to fund infrastructure because it’s boring, but enabling Science 2.0 is the Eisenhower freeway system of the mind. And then we need to get past the legal restrictions so that we can have technologies that troll for data, make sense of it, and import it mechanically.

How is Science Commons addressing those issues?

We’re sort of the public interest lawyer to the sciences. Say you want to use a database which was generated in Europe. We come up with a data protocol, a legal tool, which says “this gets your data free to the greatest extent possible in every jurisdiction in the world that we have lawyers in” (and we have lawyers pretty much everywhere, because a lot of really smart lawyers have volunteered to produce this high-quality tool).

We’re also attempting to show people what it might look like if you could wire together all this open stuff. We have a project called the Neuro Commons which is putting all the publicly available neurological literature and open databases together in a vast, open network that anyone can download, use, or build upon.

We’ve had high-throughput arrays, robotization, in silico studies, genetic sequencing, and the personal genome. All of these were supposed to catapult us off into a scientific revolution but didn’t. It reminds me of what people were saying about the personal computer in 1985: “This thing’s just a paperweight. What does it do for me?” The answer was, “Nothing until it’s wired together with all of the other ones.” Then suddenly you can’t imagine being without it.

Anurag Acharya: Founding Engineer, Google Scholar

Computer scientist Anurag Acharya and colleague Alex Verstak were onto something big when they took a break from building the Google Web index to focus on improving the rankings of scholarly articles within Google searches. The result of their sabbatical was Google Scholar beta. The specialized section of the larger Google search engine, which was launched in late 2004 and is now managed by a team of four people, has been transformational for enabling people to get their hands on all the world’s scholarly publications from their desktop. Acharya says the goal of Google Scholar is simple: a resource for anyone to find all scholarly literature across all disciplines, languages, and time periods.

Did your interest in creating Google Scholar stem from a need you saw in your own academic experience?

It was an experience I had as an undergraduate back in India. I grew up on the Indian Institute of Technology campus in Kharagpur. My uncle was a faculty member, and doing research was what the cool people did, at least in my head. I thought you did some work, and you wrote it up and you sent it for publication, because that’s what people do. You go to the library, you look up citations, you follow references, and you learn what you can.

If the papers don’t exist in your library, you write letters to people—this is 1985—and some fraction of them send you back their reprints. You send your own paper out for publication, and the reviews from the U.S. come back saying, “This is all very smart stuff, but you’re making this key assumption that is four years out of date.” So you’ve gone through all this effort and ultimately what you have done is not relevant because you didn’t know what was already being done.

With Google Scholar, first and foremost we make it possible for you to find the literature. Whether you can read it is a more complicated problem, but if you don’t know it exists, you have no hope.

Has it been difficult to persuade publishers to permit you to index their paid-subscription content?

Oh, yes. I started talking to publishers in 2001. We’re now indexing all the major publications, publishers, and societies, but it was a slow process. Initially the scholarly publishers didn’t believe that scholars used a lowly thing like a search engine. I’m serious. I had to convince people that researchers do use this. It was a mindset that search engines are used for casual things and not for real research. The attitudes really have changed.

If you could have some problem solved immediately, what would that be?

If I had one silver bullet I would apply it to translation. We index papers in every language that has any significant number of papers. We have a feature that allows you to find related articles, and relatedness can jump across language. All of this is trying to facilitate discovery.

A Google group has been working on a translation feature for many years now. There are groups that are using it to point to open-access journals and outside the English-speaking countries to make it possible for people to read papers that are not originally in English. Translation could open up the space to a population that previously we have not had an opportunity to reach.

Timo Hannay: Publishing Director, Web Publishing, Nature Publishing Group

A doctor of neurophysiology based in London, Timo Hannay manages Nature.com, Naturejobs.com, Natureevents.com, Nature Methods and Nature Protocols. He is organizer of Science Foo Camp, an annual interdisciplinary scientific “unconference” at Google headquarters. And he was a contributor to The Fourth Paradigm: Data Intensive Scientific Discovery, a collection of essays that envision the future of discovery based on data-intensive science. In the “interests” fi eld on his Nature Network profile, Hannay lists just one: “Making the most of the Web in scientific communication.”

You have called the Web “the ultimate global collaborative medium” and science “the ultimate global collaborative pursuit.”

Yes, that’s one of the reasons why I decided to work on the Web in science. Tim Berners-Lee originally considered the Web a scientific communication means. But ironically it hasn’t been scientists and the research community pushing the Web to its limits. It’s my job to try and make the Web more useful as a scientific communication medium. 

The volume of data is important and has profound implications, but an even more profound change will be if it’s all linked together. It’s going to be messy. We’re going to be using tags and microformats and ontologies and links and all sorts of strategies. But one way or another we’re integrating this data more and more. It’s not the volume of data, it’s the interconnectedness of it that’s critical in my mind.

Some would say that one of the obstacles to connecting scientific data is the traditional method of scientific publishing that doesn’t permit open access to research.

The fundamental issue is that the unit of contribution to the scientific knowledge base has become the paper. Journals grew up as a means for scientists to be able to share their discoveries and ideas. The incentive for doing so was that by publishing in journals their contributions would be recognized by citation and other means. So, you have this pact: be open with your ideas and share them through journals and you will get credit. 

Publishing in peer-review journals is no bad thing. I work for a company whose main business is publishing peer-review journals. They’re useful. However, we need to move beyond the view that peer-review publications are the only kinds of significant contributions that scientists make to the research process. A classic example would be genome sequences.

Large teams of scientists put enormous amounts of effort into providing genome sequences. Fundamentally, their contribution is making that data available to other scientists to draw insights from it. They can also provide reagents and materials to other scientists, or they can provide software and code and algorithms.

There are all kinds of ways in which scientists can contribute to the global endeavor. And yet one type of contribution, the peer-reviewed publication, has priority over all the others in the way that it’s measured and in the way that credit is assigned. The incentive structure has not caught up with what we really want scientists to do. We do want them to be able to share their ideas and their data and their reagents and so forth as well as publish traditional peer-review research reports. 

At Nature Publishing Group we try to be open to new ideas and try them out. From making tagging of scientific information possible to things like Nature Network and Nature Precedings which are venues for scientists to be able to share information with one another more informally and more immediately than they could through a scientific journal. Some things worked well and some didn’t, but that’s the nature of trying to understand a new medium and how it can be harnessed to best effect. I think the only way to effect change is by the funders, publishers, the scientists all working together.

John Wilbanks: Executive Director, Science Commons

John Wilbanks was named one of “50 visionaries who are changing your world” by the Utne Reader, and a “Revolutionary Mind of 2008” by Seed Magazine. He writes the Common Knowledge blog on Science Blogs and is known simply as Wilbanks on Twitter. As VP for Science at Creative Commons, he runs Science Commons from an office at MIT. Wilbanks joined Creative Commons from a Fellowship at the World Wide Web Consortium in Semantic Web for Life Sciences. Previously, he founded and led to acquisition the bioinformatics company Incellico.

How is Science Commons different from Creative Commons?

The primary way that we convey scientific knowledge is to compress it down into text and distribute that through a journal. But with the Internet we can now distribute a lot of the tools, data, stem cells, and so forth that used to simply be described in the paper. Making data useful to people who didn’t generate it is the most important problem, and it requires an enormous investment of time, infrastructure, curation, data standards, standard formats, and giant computers that can store it. If you add the law to that complexity, you have what we would call an NP-hard problem. Unsolvable.

When we got into this, we thought that the way we license software or literature was going to be the solution—that a Creative Commons license would take care of the problem. But data is much more foundational than literature or software and it’s more like the Web than it is like software. In other words, we all take software and run it, but the human genome is the knowledge equivalent of the Internet—it’s the common language of biotech, and if that foundational architecture imposed down-stream restrictions it would really screw things up.

The genome being in the public domain was much better than the genome being licensed. Imagine if every time a distributed annotation server ran across the genome it had to attribute whoever put that piece of genome online?

What we use instead of the law there is citation. You know that if someone published the first paper about that piece of the genome, when you write your paper you should add a citation to it. Citation norms scaled much better than the legal aspect of licensing. So, we stopped working on licensing for data and we started working on public domain pools for data. We worked on a tool called CC0—Creative Commons zero—which is a legal tool that achieves a legal status that is similar to the public domain. The idea is to waive the rights that are associated with data.

Let’s say you and I try to generate something like the genome now. If we had the money, we could sequence both of our genomes in a couple of days. But there are little bits of copyright that stick around data when you put them into a database in the U.S. They attach not to the data itself, but to the look and feel and the structure of the database. It’s unclear to many people where those rights stop and start, so the first thing CC0 does is waive those elements. 

If we want to have that data be interoperable with the public genome, we have to get rid of the database rights and the copyrightable pieces of it. The second thing CC0 does is get rid of those database rights. If we can make things legally interoperable, then the only problems we leave are the monstrously complicated technical and semantic ones. 

So, we wrote the Science Commons Protocol for Implementing Open Access Data. The first two requirements are: waive your intellectual property rights to the extent they exist, and don’t put a contract on your data. The third requirement is to request behavior through norms, not through the law. That’s about using citation, not attribution. In science, citation scales in a way that attribution doesn’t, because attribution is tied to this very old way of thinking about copyrightable object as opposed to massive data structures.

What would be one change you’d put at the top of your wish list?

It would be for the various funding agencies to put meaningful requirements or evaluation systems in place for sharing data and tools, not just papers. Right now, there’s no incentive to go through the effort of curating, annotating, and posting your data. The biggest thing the NIH could do would be to begin looking at a two-pronged mandate, similar to the open-access literature mandate, and provide minimum requirements for sharing data that you generate.

That would create incentives for researchers to get their data online and share their tools and it would create an environment where some of the startups can have success. In the absence of putting some teeth behind those requirements, all we’re going to see is an increase in the number of PDFs deposited, and I don’t think that revolutionizes scholarly communication.

Stewart Wills: Online Editor, Science

On his Twitter profile, Stewart Wills describes himself as the “aging online editor of a scientific journal, trying to stay young in 140 characters or less.” An earlier adopter than many a “digital native,” he’s been Tweeting diligently, usually several times a day, since June 2008 about all things science and media. In 2000, when he completed a PhD in geological sciences at Columbia University, Wills joined Science. His principal goal at present, he says (via his Linked In profile), is, “Keeping the Science site moving forward, to provide the best possible value and utility to users, the scientific community, and the public.”

How is your publication responding to the move of science onto the Web?

At Science we pay a lot of attention to how the digital natives are changing everything. We have a set of users with new expectations, new assumptions, new ways of learning that we in publishing need to figure out how to address. As an editor working with a scientific publication, I have an interest in making our content as available as possible and serving the community as well as possible. Whatever the business models we’re dealing with, we have to find a way to serve the community on this.

Moreso than the general population, scientists are do-it-yourselfers. If there’s a tool available, they figure out how to use it. The Web is one huge, highly flexible tool. Certain groups of scientists are in there using open notebook science and open wetware and various things like that to do their jobs. They are exploring new ways of doing science. For that reason, we increasingly hear the community’s need not just for open access but for open science—for open data.

How are you changing the way Science makes research and data available? 

The data supporting the papers has always been free on our Web site, and Science has had full text on the Web since 1996. Now we’re doing some of the more obvious things to improve the syndication of research results—RSS feeds, Twitter, and Facebook. We’re active on these social channels because that’s where the users are having conversations. It’s a way to capture some of the conversation around our content. And we are experimenting with adding different kinds of content, such as a pilot with the Journal of Visualized Experiments to create video methods to go along with certain papers. 

Would you say that scientists who aren’t on Facebook or following Twitter are at a competitive disadvantage?

That’s an interesting question and I’ll answer it this way: It’s going to depend on the network that you’re following. I heard Cameron Neylon, a senior scientist with the U.K.’s Science and Technology Facilities Council, speak at a conference recently. He filters his content through a tool called FriendFeed. It’s the most sophisticated use of tools like Twitter or Facebook to deal with the information glut: a collection of friends he trusts helps him with discovery by filtering papers that are of interest to him. It’s a certain kind of peer review.

This story originally appeared in the Spring 2010 issue of The New York Academy of Sciences Magazine.