international conference april 27-29, 2007 mit
Collaboration and Collective Intelligence
Friday, April 27, 2007
"Collective Intelligence" and "the wisdom of crowds" have become central buzz phrases in recent discussions of networked culture. But what do they really mean? What do we know about the new forms of collaboration that is emerging as people work together across geographic distances online? Are we working, learning, socializing, creating, consuming, and playing in new ways as a result of the emergence of our participation in online communities? What have we learned over the past decade that may help us to design more powerful communities in the real world? What lessons can we carry from our Second Lives into our First?
Mizuko (Mimi) Ito is a cultural anthropologist of technology use focusing on children and youth’s changing relationships to media and communications. She has been conducting ongoing research on kids’ technoculture in Japan and the US, and is co-editor of Personal, Portable, Pedestrian: Mobile Phones in Japanese Life. She is a research scientist at the USC Annenberg Center for Communication and a visiting associate professor at Keio University in Japan. email
Cory Ondrejka is the chief technology officer at Linden Lab where he leads the team developing Second Life. He also spearheaded the decision to allow users to retain the IP rights to their creations and helped craft Linden's virtual real estate policy. While an officer in the United States Navy, he worked at the National Security Agency and graduated from the Navy Nuclear Power School. email
Trebor Scholz is assistant professor and researcher in the Department of Media Study at the State University of New York at Buffalo and research fellow at the Hochschule fuer Kunst und Gestaltung, Zurich. He is founder of the Institute for Distributed Creativity and has contributed essays to several books, journals, and periodicals and co-edited The Art of Free Cooperation forthcoming with Autonomedia (NYC). email
Thomas W. Malone is the Patrick J. McGovern Professor of Management at the MIT Sloan School of Management. He is also the founder and director of the MIT Center for Collective Intelligence and author of the book The Future of Work. Malone has published over 75 articles, research papers, and book chapters and is an inventor with 11 patents. email
By Huma Yusuf, CMS ‘08
[This is an edited summary, not a verbatim transcript.]
Thomas Malone started the session by asking the crowd some questions. He first asked how many audience members had looked at Wikipedia, and whether anyone had edited content on that site. He also asked how many people had conducted an online search using Google and whether anyone had created a webpage or content that was uploaded to a website.
Malone then pointed out that anyone who replied in the affirmative to the second parts of his questions had participated in collective intelligence, while those who replied yes to the first parts of his questions had taken advantage of it.
Malone defined collective intelligence as “groups of individuals acting collectively in ways that seem intelligent,” and pointed out that by this definition, there has been collective intelligence as long as there have been humans. He said that all families, tribes, teams, and governments, for example, act with some degree of collective intelligence. In the last few years, however, phenomena such as Google, EBay, and Wikipedia have enabled a new kind of collective intelligence at a scale and in a manner that was impossible before the Internet.
Malone suggested that new versions of collective intelligence require a deeper understanding and explained that the core research question at MIT’s Center for Collective Intelligence is: how can people and computers be connected so that collectively they act more intelligently than any person, group or computer has ever done before? He then pointed to two examples of successful collective intelligence.
The NASA Clickworkers project enabled volunteers to help planetary scientists count the number of craters on the surfaces of planets being studied. When NASA initially made pictures of the surface of Mars available online, 80,000 volunteers identified over two million craters. Interestingly, volunteer ratings were almost as accurate as those by professional scientists.
Similarly, Garry Kasparov versus the World, an online chess game, utilized the collective intelligence of thousands of chess players to compete against the world champion. ‘The World’ could watch Kasparov’s moves, discuss strategy on an online message board, and determine the next move by majority vote. Although Kasparov eventually won the game in 62 moves, he said it was the hardest game he ever played, and possibly the best game in the history of chess.
Trebor Scholz presented “What the MySpace Generation Should Know about Working for Free.” He cited the member statistics for social networking sites such as Facebook (80 million) and MySpace (150 million) to indicate that within the broader context of collective intelligence, social activities take up most of the time spent online. Scholz was interested in the politics of labor on social networking sites. He defined labor much like Paolo Virno, who suggests that labor no longer has to produce an object, but can be a virtuosic performance, the act of being a speaker. Sholz explained that he considered uploading a video to YouTube or posting comments on MySpace to be acts of labor.
Scholz then described user activities online – commenting, tagging, ranking, reading, gossiping, chatting, etc. – and suggested that social life has been commercialized. He recalled the example of librarian Harriet Klausner, who reads two books a day and has written over 13,000 book reviews -- Scholz said -- for Amazon.com without receiving any compensation for her labor. Scholz emphasized, though, that since Klausner gets pleasure from her activity, this situation is not exploitative.
He then considered the economics of social networking and media-sharing sites. He reminded the audience that MySpace was bought by News Corporation for $580 million, and that the site’s projected value in three years is $15 billion. YouTube was purchased by Google for $1.6 billion and online virtual worlds are collectively worth $10 billion. Scholz said that such high valuations are commonly explained away by saying that it is an expensive, arduous process requiring a huge infrastructure to support social life online. He instead emphasized that we focus on the imbalance between what is invested in social networking sites and what content is subsequently taken in by these sites. He also debunked the other argument that it takes big business to support social life online by pointing out that the most successful sites were started by students. After addressing these arguments, Scholz concluded that the few benefit from the many, and that offline capitalistic structures are being replicated online.
Scholz also addressed the problem of centrality on the web. He pointed out that 40 percent of all online activity unfolded on 10 main websites, with 60 percent of all online content being user-generated, and 40 percent of online content being produced by big corporations. With regards to the 10 main websites, Scholz said that there is a move to ultra-concentrated web ownership, and cited the example of the News Corp monopoly. After all, many are concerned with the decision to stream Fox News into MySpace to introduce revenue for the news business given that 12 percent of all the time spent by Americans online is on MySpace. Scholz said that online monopolies were being amplified by the connection between social networking sites and telecommunication companies.
He then ran down the ways in which big businesses use online social life to their advantage. He introduced the concept of "click economies," where advertising is attracted by the number of eyeballs on a website. Social networking sites also make use of user-generated content. Facebook, for example, owns all the content uploaded to the site. Meanwhile, online user profiles are the dream of market researchers. Finally, simple transactions can also be conduced through social sites, as music can be bought on MySpace, and fashion accessories on EBay. Spam also made the list of online economic activities.
Scholz then pointed out that there is a paradox of labor on social networking sites. People who contribute to the site are not compensated directly, but they make friends, share life experiences, archive memories, trawl for jobs, find dates, become famous, “egocast”, indulge in the pleasure of creation and derive social enjoyment.
Still, Scholz pointed to some downsides of participating in online social life. He quoted from Lawrence Lessig’s Code: v2 to highlight how strong ideologies have taken hold online. Facebook, for example, only gives students the opportunity to identify themselves as male or female on their online profiles, even if they consider themselves transgender. Scholz also referred to the fine print of social networking sites such as Facebook and pointed out how personal information uploaded to the site is aggregated to make databases.
Scholz then discounted what he called the neo-liberal argument that people are always free to leave social networking sites. He instead argued that people cannot leave such sites because their community and content exist there. He also suggested that the fact that 700,000 people protested Facebook.com’s decision to include a news feed proved that they couldn’t leave the website even when unhappy with it because their friends and social life remained online. Scholz suggested that it was for these reasons that intellectual property issues were starting to come up with regard to social sites. He also advocated for hybrid, non-profit alternatives to sociable media giants.
In conclusion, Scholz argued that net publics should have full control of their online public and media companies should pay attention to exit costs and make it easier for people to leave websites. He also suggested that social sites increase transparency by emphasizing what is currently in fine print so that people know exactly what happens to the profiles and content they upload. He also argued that people should get a fair share of the money that they help generate. He recognized that to make a new system work, people would need increased media literacy.
Cory Ondrejka began his presentation by connecting the virtual world Second Life with moderator Thomas Malone’s The Future of Work. He explained that Second Life had been operational for less than a year when Malone’s book was released. The book helped Linden Lab – with less than 15 employees at the time – make decisions about their evolving business model, and inspired them to opt for a radically decentralized business model.
Ondrejka then explained that Second Life is not a game, nor a website. It is a virtual world created by its users, a platform that provides users with the software and technologies they need to build a virtual world. Anything people build in Second Life is theirs to own, sell, keep, transfer, move around, or copyright. Linden Lab does not claim intellectual property rights to anything created in Second Life. Unlike video games, the virtual world has no artificial conflict or sense of outcome.
Ondrejka further explained that Second Life has six million registered users, with an excess of four million people actively logging into the virtual world. About 650 square kilometers, an area roughly the size of Singapore, have been simulated in Second Life. In terms of economy, Second Life generates $60 million US in GDP a month and boasts a foreign exchange of $8 million US (as calculated in March 2007). Second Life’s currency is called the Linden Dollar, and is exchanged for US dollars and vice versa.
Ondrejka then described how collective intelligence and peer production influence activities within Second Life. Firstly, he pointed out that the Linden Lab can aggregate the value of creations in Second Life – which is how they arrived at the $60 million US figure – as well as the time spent by users in the virtual world. When aggregated, over 340,000 hours a day are spent in Second Life, with 20 percent of the time spent building things. That amounts to 34 user years per day of content creation by what divides out to a 13,000-person content creation team. Ondrejka pointed out that if he had to pay such a large team, it would cost Linden Lab about $1.3 billion US a year.
Instead, people pay to spend time and build things in Second Life.
And they often spend their money and time collaborating with others in the virtual world.
The difference between experiencing an online virtual world and the more general experience of using the Internet, Ondrejka said, is that using the web remains a primarily solo experience. He described blogging as a involving one person climbing to the top with a megaphone and yelling out his opinions until he is displaced by another person with a megaphone. Ondrejka argued that this was not a valid form of communication or community and instead suggested that Second Life, which fosters more collectivity and conversation, was a better example of community.
Regarding collective intelligence, Ondrejka cited some examples of how users have collaborated in Second Life.
The first example of virtual collaboration Ondrejka gave was Noah, a project built in the virtual world to facilitate the collaborative consumption of weather information across the globe. People from around the world could describe weather patterns in their area and gain insights into weather trends.
A second example concerned the Starwood Hotels’ virtual Aloft Hotel in Second Life, launched to generate press attention for the company. Once people began visiting the virtual hotel, they also began complaining about and critiquing the hotel’s design. Starwood then began surveying people who visited their virtual hotel and generating aggregate data to determine how people were moving through the space. Eventually, Starwood announced that they would be renovating and redesigning their real-life hotels in response to feedback received via Second Life. In that sense, Ondrejka argued, non-experts were able to participate in a collective creative act and a design process and influence the building of a hotel.
In fact, Ondrejka said, Second Life users work collectively to give Linden Lab negative feedback about the product. In fact, there’s enough of a desire among users to give feedback that a virtual cottage industry making animated protest signs exists within Second Life. Targeting Second Life’s initial economic structure, for example, protestors congregated at sites where new users were arriving and slowed traffic to the point that they got the attention of the Linden Lab employees. As a result, Second Life changed the way taxation worked in the virtual world.
Mimi Ito addressed the topic of collective imagination in the digital age and discussed her ethnographic work with young fans of Japanese animation and games. She pointed out that children are growing up in new media environments where an abundance of media references are mobilized in various ways to create a collective imagination. With the advent of amateur and digital production, the possibility of customizing and personalizing one’s relationship to media content exists. Moreover, Ito argued that media is becoming a conduit through which we link to a shared body of knowledge and culture that exceeds the grasp of the individual.
Ito suggested that the individual mobilization of a collective imagination is the way young people conduct their everyday social transactions and communications. She acknowledged that this mode of participation has existed for as long as there have been stories and music to share, but emphasized the qualitative shift in the forms of media people can use such as Tamagotchi or YouTube.com. Ito pointed out that rich media content is the vocabulary through which we traffic, share, and communicate with people, tell them who we are and what our interets are. Ito called this process hypersociality, a sociality that is augmented by a dense set of media signifiers.
Ito used the example of Pokemon to further illustrate her point. She said that in the new media environment in which children are growing up, an abundance of new media references is complemented by an abundance of new media forms, for instance, narrative media forms and digital media forms that are facilitated by and distributed over the Internet. Ito explained that in Japan, the plurality of media forms, of which Pokemon is a perfect example, is referred to as the “media mix.” Japanese media mixes – portable mobile phones and other devices – are central drivers of new media usage. These media mixes show how practices of engaging and sharing are migrating out of stationary locations to a wide variety of settings in everyday life.
In this context, Ito suggested that Pokemon was a breakthrough media form in many ways, partially because it placed mobile media – Game Boy games and trading cards – at the center of the equation. More importantly, Pokemon boosted the complexity of form and content of children’s media to unprecedented levels.
Until Pokemon, people assumed that children’s media needed a few characters and simple narrative content. But Pokemon demonstrated that kids could process hundreds of characters, each with unique characteristics. Through Pokemon’s success, the industry learned that social exchange and hypersociality could make content widely popular.
Ito pointed out that Pokemon is important from a learning perspective and in the context of collective imaginations because it involves the intense exchange of information as well as an exchange of physical media such as trading cards. As they play with Pokemon, children realize that they’re participating in a collective imagination that is greater than what they could master on their own. Ito added that such examples are bolstering the new premise that the best learning will happen in groups and social settings. Ito admitted that these everyday exchanges between children may not seem as spectacular as efforts such as Wikipedia.com, but argued that the broad base of participation by children would eventually drive new media systems.
At the end of the presentations, Malone pointed out that while Scholz had emphasized the bad aspects and dubious ethics of big media, Ondrejka had emphasized positive developments in Second Life and Ito celebrated collective imagination in children’s culture. He challenged each panelist to take a stance opposing the one taken in their presentations, meaning that Scholz would have to highlight positive aspects of MySpace, while Ondrejka and Ito would have to imagine downsides to collective intelligence.
Scholz first said that he had not made an extreme anti-business argument. He acknowledged that people enjoy the convenience and simplicity of MySpace, particularly its features that are designed to support sociality at a mass scale. But he reiterated that the Internet was becoming centralized. He said that a few years ago, the web was decentralized, replete with people experimenting with code and being creative. Now, however, there is a centralization of the web enabled by monopolies. Core sites have immense power that can be easily abused as a result.
Ondrejka discussed the problems posed for Linden Lab by collective and distributed intelligence. He explained that the lab’s internal development culture that applied to 140 employees in five locations was to choose tasks wisely, and then execute them well. He admitted that giving people the power to research, choose, and execute what they thought was important in the absence of a manager assigning work meant that it was harder to address broader issues. The work culture at Linden Lab made it harder, but not impossible, to change direction, think on a global scale, or address the public good.
Ito responded to Malone’s question by discussing the moral panic surrounding Pokemon, even though she does not agree with it. She said the mainstream view is that Pokemon is radically bad for children as it is nothing more than an uninteresting and useless media form. But she suggested that adults who tried playing the Pokemon card game would learn that there are complexities in the game that aren’t visible to outsiders. She elaborated that the moral panic around Pokemon feeds into deep-seated fears about what’s appropriate for children. Pokemon gives children autonomy, independence, and a taste of financial entrepreneurship that has been deemed inappropriate for children. Thus, people find it threatening when children are actively engaged in media forms and participating in entrepreneurial narratives.
Ondrejka responded to Ito’s comments by highlighting the inevitable amateur-expert collision that collective intelligence entails. He explained that amateur should be understood as “lacking credentials,” rather than as ill-equipped to participate. He pointed out that one of the most interesting things about Second Life was seeing how amateurs were entering fields such as therapy and neurology in the virtual world, coming up with interesting ideas, and running into, and up against, professionals.
Scholz suggested that the content of what people say no longer matters. The fact that so many people now experience themselves as speakers is what is important and that phenomenon will lead to a more participatory culture.
An audio recording of Collaboration and Collective Intelligence is now available.
In order to listen to the archived audiocast, you can install RealOne Player. A free download is available at http://www.real.com/realone/index.html .
A podcast of Collaboration and Collective Intelligence is now available from Comparative Media Studies.
A webcast of Collaboration and Collective Intelligence is available.