This article was taken from the June 2011 issue of Wired magazine. Be the first to read Wired's articles in print before they're posted online, and get your hands on loads of additional content by subscribing online.
The scrobbling technology used by Last.fm -- as well as some of the $280 million exit capital that the site's backers received -- has given rise to Mendeley, a software platform that intends to be the most disruptive force in the history of academic publishing.
I. Introduction
Walk through the ground floor of German entrepreneur Stefan Glaenzer's White Bear Yard in Clerkenwell, east London, and you see an apparently standard tech hub: open-plan office, exposed brick walls, coders staring at screens, armchairs dotted about for lounging. The place shouts "startup". But the ambitions of those in the room extend far beyond just building a business. Rather, they tell you sincerely that they are attempting to change the world. Even if this revolution will be spurred by an organisational tool for academics.
Victor Henning (profiled in 03.10), a boyish German, is the 31-year-old cofounder and CEO of Mendeley, whose software platform extracts research and aggregates information that users put into their Mendeley libraries in the form of PDFs, Word documents, Excel or Photoshop files, or websites of academic journals and research papers. Mendeley Desktop manages the information. Mendeley Web indexes external Open Access databases and repositories, finds data online and provides an interactive platform for researchers to exchange insights and blog progress. Sounds arcane? It's booming. "Currently 800,000 researchers from 195 countries and 19,150 universities, research institutions and NGOs have uploaded seven million research papers to our site," says Henning. "It helps researchers and scientists work smarter."
A fuller explanation comes from Jeffrey Lancaster, a PhD student in photochemistry at New York's Columbia University and a Mendeley convert. "It organises what you have," he enthuses, "and pushes feedback about your own data, priorities and even outstanding questions based on the literature you have in your repository. It fills potential knowledge gaps with crowdsourced citations by recognising that [which] a researcher may not be aware of. It can fundamentally change the way we ask and answer important questions in many disciplines." But can it disrupt a publishing industry where the biggest players are powerful and entrenched media companies such as Reed Elsevier, whose 2010 revenues were over £6 billion?
The last day of February is hackday at Mendeley, a time for the team in the London office to voice new ideas. Whiteboards bear the results. "I think this one," says Henning, struggling to decipher the marker-pen hieroglyphics, "is for a WordPress blogging platform. A website that gives you related citations. It scans blog posts, extracts words and finds which research matches. And this one...well, I've no idea what that's about."
Henning seems happily diffident, as though demonstrating something interesting that he created but has now handed on, something everyone else is free to play with and improve, by way of
crowdsourced apps. But although the atmosphere in the office is relaxed and low-key, what Henning and his group are up to is boldly innovative. The Mendeley team is attempting nothing less than changing the way scientists conduct and share their research.
II. Early Days
In 2008, Timo Hannay, then head of web publishing at the Nature Publishing Group and responsible for managing onlineresearch aids such as Nature Network and Connotea, mused on his company website about its biggest future threat: "I think it is the unknown grad student in his (or her) dorm room hatching a plan to turn scientific communication upside down in the same way that Napster, Google and Wikipedia disrupted other industries."
Henning was that grad student. His dorm was at the elite WHU Otto Beisheim School of Management in Vallendar near Koblenz, Germany. By his own admission Henning was an unlikely-look ng business student. "I was a punk," he says. "I had a T-shirt with
'Anarchist' written on it, my hair was spiky and I had a pierced lip. My father stopped talking to me because of the way I looked, and on my first day at college someone came up to me and said, 'Ah, a rebel. Every class has one.'"
Sure enough, Henning found he had something to rebel against at WHU. The time it takes a research student from finishing a PhD, through peer review to publishing is at least two years, often longer - an aeon in scientific research and in an academic career.
To Henning, the process seemed unnecessarily and maddeningly drawn out. One of the strongest checks on progress is the need for citations, the references in PhDs and academic journals that serve as the main measurement of the worth and usefulness of research.
Successful completion of postgraduate degrees and funding for further work is a matter, in essence, of adding up how often you are mentioned in other people's work and finding the correct citations for your own work. Mendeley potentially cuts through that entire process, most obviously by vastly simplifying the search for citations.
No wonder some scholars see Mendeley as a huge step forward for science. "I was a pretty heavy user of similar technologies before," says Rick West, assistant professor in education technology at Brigham Young University in Utah. "I had used citation-management
tools that allowed for collaboration and sharing through the web, for example. However, the problem was that I needed a couple of tools to do what Mendeley does for me now. It has made it much easier to do my research, because I can save citations, organise them into folders, share them with my collaborators, and annotate the PDFs all in one place. I like how I can work offline and online, and I love how I can share my annotations and notes with others." "I don't like waiting," Henning says. "Actually, I hate waiting -- I want things to happen now. And I thought that there must be a technological solution that would make the process easier." Fellow student Jan Reichelt, now 31, felt the same way. After both students collaborated on a case study for an entrepreneurship course that Glaenzer was giving at WHU in 2003 (which, in turn, became a book about Europe's fastest-growing companies), they began to think about developing a model: in essence, Last.fm for scientists. But instead of scrobbling MP3s, it would scrobble PDFs.
Henning and Reichelt - a keen salsa dancer, clean-cut and controlled in his mannerisms - were then joined by Paul Föckler, who had also been a postgrad at WHU. By chance they had met him one night while on a trip to Las Vegas.
Soon, the three of them were aggressively focused on building their vast scientific pool. By January 2010, Mendeley had real-time data on 28 million research papers and was being used by academics at some of the top research universities in the UK and US, including Stanford, MIT, Cambridge and Oxford.
The hard work needed to create such disruptive technology came naturally for Henning: his Korean mother ran a newspaper-and-tobacco stand in Germany, and his father had worked his way from mechanic to motor-industry executive. Each of the three founders worked long hours while writing PhDs and holding down jobs. At the beginning meetings were conducted on Sunday nights from 11pm to 3am via Skype. Henning recalls arguments. "I would lose my temper and yell, but that's the beauty of Skype, it doesn't go further than that," he says. But as well as passion there was complete trust between the three, and from the start they spilt duties equally. "I don't like waiting," Henning says. "Actually, I hate waiting -- I want things to happen now. And I thought that there must be a technological solution that would make the process easier." Fellow student Jan Reichelt, now 31, felt the same way. After both students collaborated on a case study for an entrepreneurship course that Glaenzer was giving at WHU in 2003 (which, in turn, became a book about Europe's fastest-growing companies), they began to think about developing a model: in essence, Last.fm for scientists. But instead of scrobbling MP3s, it would scrobble PDFs.
Henning and Reichelt -- a keen salsa dancer, clean-cut and controlled in his mannerisms -- were then joined by Paul Föckler, who had also been a postgrad at WHU. By chance they had met him one night while on a trip to Las Vegas.
Soon, the three of them were aggressively focused on building their vast scientific pool. By January 2010, Mendeley had real-time data on 28 million research papers and was being used by academics at some of the top research universities in the UK and US, including Stanford, MIT, Cambridge and Oxford.
The hard work needed to create such disruptive technology came naturally for Henning: his Korean mother ran a newspaper-and-tobacco stand in Germany, and his father had worked his way from mechanic to motor-industry executive. Each of the three founders worked long hours while writing PhDs and holding down jobs. At the beginning meetings were conducted on Sunday nights from 11pm to 3am via Skype. Henning recalls arguments. "I would lose my temper and yell, but that's the beauty of Skype, it doesn't go further than that," he says. But as well as passion there was complete trust between the three, and from the start they spilt duties equally. It's the same split they operate today: Henning fronting the business; Reichelt managing it -- he travelled to Belarus to work with the outsourcing company on the first protoype; and Föckler, the quietest of the group, ensuring the system functions optimally.
All three founders are at the hackday. Reichelt is in town from the New York office that Mendeley opened in September 2010. He's upbeat and confident through he bemoans the stateside salsa scene. "I like traditional Cuban salsa," he says, "In New York, it's mambo on the two-step." Föckler, as befits the nuts-and-bolds man, is a quieter, understated figure yet no less driven. He offers an illustration of the good that Mendeley could potentially do. "Say a woman gets a rare kind of skin cancer," he says. "She can find out who is doing the best research, who is taking on patients, which institutions are making breakthrough, and can join together with other patients on a Mendeley group and share knowledge and experience."
The excitement of such potential outcomes runs through the very core of a company that takes a remark by Tim Berners-Lee at TED2009 in Long Beach as its founding statement: "A lot of the state of knowledge of the human race is sitting in the scientists' computers and is currently not shared...We need to get it unlocked so we can tackle those huge problems." Ian Mulvany, vice-president of new product development -- poached in 2010 from Nature -- says: "People here have a real understanding of what we are doing. I believe in science and in its ability to make the world a better place and answer our questions. I'm passionate about that."
His colleague Jason Hoyt, Mendeley's vice-president of R&D and chief scientist, nails the Mendeley mission in ten words: "Trying to create a balanced business and change the world."
III. AWARDS
The first suggestion that something special was happening came at Plugg in Brussels in 2009, a one-day conference that celebrates entrepreneurship and innovation. Henning stayed up the night before to work on his presentation and he hadn't eaten by the time he took the rostrum. Shaking with hunger and slightly disheartened by the slickness of the other entries, he thought Mendeley had blown it when he stepped down.
Far from it: Mendeley won the Startup of the Year award. To celebrate, Henning went out for "a good French steak and a bottle of wine". Which, despite his anarchist history, seems to be as wild as Henning allows himself to get now that he is a CEO. More recognition followed, including Best Social Innovation Which Benefits Society at the 2009 TechCrunch Europe Awards and Startup Most Likely to Change the World for the Better at The Guardian Activate Summit 2010.
The key moment, though, was getting Glaenzer -- the investor best known for Last.fm -- to sign up in 2009. With him came Alejandro Zubillaga, formerly of Warner Music Group, and Ambient Sound Investments -- set up by four founding Skype engineers -- as well as a combined $2 million of development funds. Today, Glaenzer operates on the first floor of the London base he has set up with the ambition of creating European online companies that can take on the best in the world. He is near enough to Henning and the team to keep an eye on them but not be on their backs. "When they pitched Mendeley in 2007," Glaenzer enthuses, "I knew it was good. Jan, Victor and Paul had incredible passion. There are several types of entrepreneur out there. Some are in it for the money, but these guys just want to deliver a way better service."
Online scientific databases are not an entirely new idea. There are already online and desktop research aids such as Zotero, Papers and CiteULike. "Yes, but [Mendeley] is about the tweak that allows automatic recognition of the metadata in combination with the scrobbling of the articles," Glaenzer says. "Think of Google.
Search engines were out there before Google. It is beneficial for society, it's a huge market and it's a market that has not been digitally disrupted so far, so there was room for play in that.
There is great pleasure in that disrupting."
Glaenzer may be excited by Mendeley's potential to disrupt the traditional academic-journal and research model, but other figures in the industry are pointing to a future that is about co-operation as well as competition.
Rafael Sidi, vice-president of science direct at Elsevier Online, is responsible for the SciVerse app gallery. Containing an Elsevier-developed Mendeley app, it pulls Mendeley readership stats into SciVerse. "The reason that I approached Mendeley is that I saw something very valuable and very interesting that they are providing our customers, so our customers are already using their product, accessing Mendeley from my system," he says. "If, as a 21st-century business, we are not enabling scientists and researchers to come up with better outcomes, then we are not serving our community. If I don't have Mendeley services but my customers are using them, then I need to make my platform easier for them to access the Mendeley service. I cannot say, 'Because I haven't built this, I do not want to provide access to them.' If you look at the core of what publishers are doing with these applications, they want to better the world."
Timo Hannay, now head of Macmillan's new digital-science division, talked to Henning about the opportunities technology is offering academic publishing. "Generally," he says, "the most effective way to get scientists to engage is to offer a piece of functionality that meets the individual and selfish needs of the scientists themselves." Mendeley Desktop provides just that, but Hannay is less starry-eyed about collaboration. "That can be a hard sell to scientists: 'You need a tool that helps you share things.'
Why would I want that? It's easy to be naïve and you have to be nuanced, but ultimately it's about helping scientists to get the job done. About making the next discovery."
And revenue? Glaenzer, who is funding much of the project (he and several other backers have invested in the region of £3.5 million so far), is sanguine. "It's not at the stage where we monetise our community yet; we're getting as much knowledge and usage as possible. Mendeley is getting four or five unsolicited emails a week from big corporations about whether or not we can produce an enterprise version behind firewalls. But we are just not focusing on that.
These guys should work on the core priorities of Mendeley."
Henning, however, is more clear about the business model. He explains that a) the productivity/collaboration component of Mendeley will be monetised, with accounts being rolled out later this year; b) the unique data that's aggregated will be monetised; c) Mendeley will be turned into a content distribution platform; and d) precisely targeted advertising will be aimed at Mendeley's highly specialised audience. "In principle, those revenue lines are sensible," Hannay observes. "No one really knows just how successful they'll be for this kind of application, but lots of people have developed similar software -- Papers, Sente, Qiqqa, Colwiz, Peaya Paper, WizFolio, Pubget, DeepDyve etc -- based on one or more of these potential revenue lines. In that sense, they're standard. I guess my main observation about Mendeley is that they seem to have spent a lot of money before even trying to monetise the application."
If there were to be an influx of cash it is more likely to come from a corporate buy-out, no matter how idealistic the team.
Glaenzer's exit from Last.fm -- sold to CBS in 2007 -- was worth $280 million to him and the other backers. Does he fear that his protégés at Mendeley will be similarly tempted? "That is the downside of being an investor -- because this call is finally a call for the founders on whether or not they want to sell," he says. But Henning doesn't appear to be interested in selling; he doesn't even talk about the business side of Mendeley a great deal. "You know, even with the anarchist T-shirt and the pierced lip, I think I was actually more conservative when I was at WHU, more into making money than I am now. Now I care about what I can do, about what we can do."
Besides, there is an enormous challenge: "We are aiming to make Mendeley the biggest knowledge database on the planet," Glaenzer declares. "In 19 months, we have collected over 67 million articles. It took Thomson Reuters 49 years to come up with 40 million. But in the end you spend a lot of time with the work, and if you can have some impact on the world and you can make life better? That's as rewarding as it gets."
IV. REVOLUTION
Any endeavour based on scientific research might, therefore, become that much easier because Mendeley made the results available more quickly, or just joined the dots between research approaches that might previously have gone unrecognised beyond their immediate forum. "Mendeley seems driven by the goal of more openness in science and the benefits that can come from more openness," says Jonathan Eisen, professor in bioinformatics, University of California, Davis, and PLoS Biology's academic editor-in-chief. "In a way they are saying,
'What new things can we develop that both encourage and take advantage of more openness in sharing publications and related materials?' "This is revolutionary because many of the other systems really seem focused on just one main concept, like being an online reference-management system or a place to comment on articles or such. Mendeley is more of a platform for sharing and for making use of the information that comes from sharing. By being a platform driven by sharing, they seem to be the place that will discover and develop new tools and new ways to think."
Henning walks back through the office, past Reichelt, Föckler and the other members of the team, looking both contented and excited. He's happy to be getting on with changing the world.
This article was originally published by WIRED UK
