March 31, 2008

Five principles for successful mass collaboration, part 1

Author: Charles Leadbeater

Linux has succeeded as a product only because the community that supports it has organised itself systematically to create, share, test, reject, and develop ideas in a way that flouts conventional wisdom. Successful We-Think projects are based on five key principles that were all present in Linux. Here are the first two.

This article is excerpted from the newly published book We-Think: The Power of Mass Creativity.


Everything has to start somewhere. Somebody has to be willing to work harder than everyone else or nothing ends up getting done. Innovative communities invariably start with a gift of knowledge provided by someone, just as Linux started with the kernel that Linus Torvalds slaved over and which he posted on the Internet.

A good core attracts a community of capable contributors and developers around it. The kernel has to be solid but unfinished, so open to improvement; if it were already complete there would be few opportunities to add to it. Jane McGonigal says the core to a successful game like I Love Bees depends on the starting-point being ambiguous and open to interpretation. Both the worm project to decode C. elegans' genome and I Love Bees began with a puzzle that could be solved only with the collaborative efforts of people with different skills. Steven Weber, a political scientist at Berkeley University in California, found that successful open source software projects tended to be 'multi-dimensional' and complex, thus inviting the involvement of people with different skills.' Thomas Kuhn summed up the ambiguous character of the core to a new intellectual community in his history of scientific revolutions. Kuhn argued that the possibility of a new scientific paradigm emerged when a small group of pioneers made a breakthrough that was

sufficiently unprecedented to attract an enduring group of adherents away from competing modes of scientific activity. Simultaneously, it was sufficiently open-ended to leave all sorts of problems for the redefined group of practitioners to resolve.

However, a core will develop only if its creators give away the material on which others can work, to which they can add and which they can refine. Successful innovation comes from a creative conversation between people who combine their different skills, insights and knowledge to explore a problem. We-Think is creating a new way for these conversations to emerge. A good core starts a creative conversation, and invites people to contribute.


A successful creative community has to attract the right mix of people, who have different ideas and outlooks and access to tools that enable them to contribute. We-Think takes off only by getting the right answer to each of the following questions: Who contributes? What do they contribute? Why do they do so? And how do they do it?

Creative communities have a social structure. A relatively small, committed core group tends to do most of the heavy lifting: the discussion moderators in Slashdot; the original inhabitants of Second Life. These are the Web 2.0 aristocracy: people who because they have been around longer and done more work tend to get listened to more. There is nothing unusual in this. Most innovative projects, whether inside a company or a theatre group or a laboratory, start with intense collaboration among a small group which shares a particular passion or wants to address a common problem -- as did the worm researchers who gathered around Sydney Brenner at Cambridge. Often, however, such communities can become closed and inward-looking. To be dynamic, they have to open out to a wider world of more diverse contributors who add their knowledge or challenge conventional wisdom.

We-Think projects take off when they attract a much larger crowd, who are less intensely engaged with the project. Their occasional, smaller contributions may in aggregate be as significant as the work initially done by the core. Linux, for example, as well as having 400 key programmers at the core, has close on 150,000 registered users -- akin to members -- who may only report the occasional bug in the program. Yet such a report may provide the starting-point for a much more significant effort at innovation. The make-up of the crowd is as important as the brainpower of the highly committed core. Crowds are intelligent only when their members have a range of views and enough self-confidence and independence to voice their opinions. Scott Page, a professor of complex systems at the University of Michigan, used sophisticated computer models to find that groups with diverse skills and outlooks came up with smart solutions more often than groups of very clever people who shared the same outlook and skills. Groups made up of many people who think in different ways can trump groups of people who are very bright but very alike, Page argues, so long as they are organised in the right way.

Page's explanation is that the more vantage points from which a complex problem is seen, the easier it becomes to solve. A group of experts who think in the same way is probably no better at devising a solution than just one of them, so adding more people who think in the same way is unlikely to improve a group's ability to come up with different solutions. Groups who think in the same way can often find themselves stuck at the same point -- akin to their being at the peak of a foothill in a mountain range, unable to climb to the higher peaks that lie beyond. A group who thinks in diverse ways, in contrast, is more likely to address a problem from many angles, less likely to get stuck and more likely to find a way out if it does get stuck. Diverse viewpoints are likely to generate more possible solutions and evaluate them in a wider range of ways. The right perspective can make a difficult problem seem easy. Innovation often involves trying out many vantage points before finding the one that makes the problem look simple. As Thomas Edison put it, 'We have found 1,000 ways not to make a light bulb.'

Bugs in a software program often become apparent only when the program is tested in many different settings. Better 1,000 people making different tests at the same time than a single person making 1,000 tests one after the other. This explains why open source programs are often more robust that proprietary software: they have been tested much earlier by a much wider group of users. Bart Nooteboom, a professor at Rotterdam University, argues that distributed testing of this kind is vital to most innovation. He examined the development of 17th-century Dutch sailing ships and found that the designs mutated when the community of sailors tested and then adapted them to meet different conditions: first canals, then lakes, larger inland waterways, offshore sailing, the North Sea, the Atlantic and so on. We-Think allows ideas to be tested from a larger, more diverse set of vantage points more quickly and with ideas continually passing between the tightly knit core who develop them and the crowd who test them out.

This testing becomes possible only when people can make the kind of contributions they feel happy with, which requires tools to allow them to get involved. Mass computer games thrive by making it easy for player-developers to pick up tools to create content. Blogging depends on easy-to-use software for writing and publishing online. The camera phone is now a ubiquitous tool for citizen journalism. Such tools are taking to mass scale the self-help ethic of the original computer hackers. The first versions of the Unix operating system, on which Linux is based, were created by lone programmers who could not afford to provide tech support to their users. So when they sent their programs to people, usually on a stack of floppy disks, they included a set of tools that allowed users to sort out problems themselves. When people can get hold of tools that allow them to produce aspects of a service, they start becoming players, participants and developers: newspaper readers become writers, publishers and distributors; bystanders become photographers; the audience can become reviewers and critics.

Perhaps the most perplexing question is not how people contribute, but why they do so -- particularly when they are not being paid and their work is given away. In open source software projects, a few are inspired by a hatred of proprietary software providers, especially Microsoft. A minority are driven by altruistic motives. Some see their involvement as a way to get a better job: by showing off their skills in the open source community they can enhance their chances of being employed. For the majority the main motivation is recognition: they want the acknowledgement of their peers for doing good work that they enjoy, that gives them a sense of achievement and in the process solves a problem for which other people are seeking a solution. Many of the most striking Web 2.0 success stories started when users created tools to solve a problem they themselves faced -- keeping track of all the blogs being created, sharing video and photographs online -- and which quickly got taken up by others who faced similar problems.

Open source gives away intellectual property so other people can freely use it. We-Think requires more than that: it is also an invitation to participate and collaborate in creating something. Open source ownership of a project becomes powerful when it enables mass collaborative approaches to innovation. For that to be possible many ideas have to be combined; contributors have to meet and connect with one another.



  • Collaboration
Click Here!