Advanced Search

Please click here to take a brief survey

Rheingold on Cooperation
Alex Steffen, 1 Feb 04

If you're paying even the slightest bit of attention, you've probably caught onto the idea that the coming decade is going to be one of exploding possibilities for collaboration and cooperation.

Worldchanging buddy Howard Rheingold is launching a project to understand how best to help cooperation and collaboration flourish. He's started by putting together a really sharp overview of the field (PDF, excerpted below). It's fine work, with explorations of Reed's Law, open source and social software, blogging and how to avoid the tragedy of the commons - as well as a solid introductory bibliography.

From now on, when I want to explain to someone what we talk when we talk about collaboration, I'll print this out and hand it to them.

"Commons foster innovation."

"The aggregate transformative effect of millions of people carrying and wearing super-computing power, with high-speed connectivity is creating a new threshold of social organization, an unprecedented scale of collaboration. At this threshold, we are seeing the early forms of a new literacy of cooperation. The technological components—the Internet, mobile devices, and their powerful hybrid—are in place. However, the overarching framework for a new way of thinking about cooperation does not yet exist. The knowledge component is lagging. Nevertheless, we can already begin to glimpse the outlines of such a framework in a number of different realms today."

"Q. | Who are some of the key players involved in building the new theoretical frameworks for cooperation and collective action?

"Robert Axelrod, at the University of Michigan, has combined new understandings from biology, economics, and computation. He has focused specifically on questions about the evolution of cooperation in biology by using computerized strategy games such as “Prisoner’s Dilemma.”

"Lynn Margulis, at the University of Massachusetts, has demonstrated that the early Darwinian emphasis on competition as an evolutionary engine provided only a partial explanation. Symbiosis and cooperative arrangements undergird much of what is now understood about the mechanisms of evolution.

"We should also look at what’s emerging in our understanding of armed conflict and peacemaking. Recent field work in El Salvador by Elisabeth Jean Wood, at New York University, on “political violence and robust settlements” offers evidence that both sides of the long, bitter civil war in that country unconsciously used game-theoretic strategies in their mutual withdrawal from conflict.

"Finally, in the realm of environmental policy and the political management of common resources, the work by Elinor Ostrom, at Indiana University, and others in the sociology of common pool resource management has revealed that grazing pastures, hunting grounds, and fisheries need not fall into the “tragedy of the commons.” Rather, they can be managed locally, through ad-hoc social contracts that seem to have a general resemblance across eras and cultures."

"Q: | You’ve begun a new project with the Institute for the Future to develop the literacy of cooperation. What’s your sense of the task before us?

"Our present level of knowledge about the role of cooperation and collective action in human enterprise is scarcely higher than knowledge about disease before the discovery of microorganisms.

"Descartes decreed that a “new method” was required to think about the physical world. That new method of thinking—the scientific method—led to biology, and biology created the knowledge that served as the foundation for medicine.

"Before we can approach the solution to problems of conflict, cooperation, and governance of an interconnected global world—the “medicine” for social ills, if you will—we need new fundamental knowledge. We need the equivalent of a “biology” of collective action. And for this interdisciplinary understanding to emerge, a new way of thinking across disciplinary boundaries is required. The technology of collective action provides the infrastructure for its own future evolution. Whether or not the deep understanding of cooperation can be catalyzed to knit together the separate strands of inquiry remains, however, a critical uncertainty. Success likely leads to a scenario of peer-to-peer abundance. Failure—which emphasizes control over cooperation—likely leads to political stalemate and stagnant technology."

"Connectivity has a value, and that value changes with the kind of connectivity. For example, the value of a many-to-one connection—such as a cable TV service—grows as the number of customers grows. If the value of the connection to the cable company is $10, the value of the entire service is 10 times the number of customers.

"But in a one-to-one network, like a telephone network, the value of the network grows much faster as the customer base grows: if there are two customers, they can only call each other; if there are three customers, there are eight possible connections. So the value of network grows at the rate of N2–N—or for all intents and purposes, as the square of the number of customers or nodes. This is called Metcalf’s Law, after 3Com founder Robert Metcalf, and it applies to lots of types of networks, including the Internet and local area networks (LANs) that connect devices within an organization or home. It also accounts for the rapid growth of the economy as the Internet became connected.

"Recently, David Reed, at MIT’s Media Lab, has identified a third type of network with an even greater connectivity value. He calls these groupforming networks(GFNs). These are networks that explicitly support forming affiliations among subsets of their customers. Social software, such as Ryze, Tribe, and Friendster, are examples of GFNs. Reed argues that the value of potential connectivity for transactions in these kinds of networks grows exponentially.

"Here’s his logic: Every GFN represents a certain number of possible subsets as small as two people (or nodes). So if the value of the network increases as the number of possible subsets, it increases at 2N–N–1, or approximately 2N. This potential for creating exponential growth of value is what is driving the rapid growth of social software offerings today. It is one measure of the value of collaboration."

Bookmark and Share


hmm, the link is not working and all the exp in the formula have been eliminated, for example:
the number of possible groups on a set of size N is 2^N, so the it grows as (2^N)-(2^(N-1)). Not as 2N, or 2N-N-1 ( = N-1?).
For the rest is great.

Posted by: Pietro on 1 Feb 04

Link fixed.

Posted by: Jamais Cascio on 1 Feb 04

Howard's inclusion of in his list of "social software" alongside Ryze, Friendster, et al is very curious. Unlike most of the products that are usually lumped together as "social software", MoveOn doesn't do anything to foster connections between its users... it is simply a tech-savvy advocacy organization that has done an effective job of using email for rapid-response organzing.

On a larger level, I'm not sure I really see Howard describing anything that's truly new. Collaboration and cooperation have been happening for decades -- the Internet has certainly greased the skids, but I honestly wonder if it's really changed the fundamental underlying dynamics that lead to effective (or ineffective) collaboration....

Posted by: Jon Stahl on 1 Feb 04

I'll also object to the idea of some sort of exponential growth in GFN network value. The value of the network doesn't grow as the number of *possible* subsets grows, it grows as the number of *actual* subsets grow. A single person can only be in so many of the subsets, which limits the total number right there. Adding a new person to an existing network increases it's value by the number of groups that person is capable of meanfully taking part in, a (relatively) constant value. How many usenet groups can you read and contribute to? That number certainly doesn't increase as the number of usenet groups increase.

Posted by: Craig on 2 Feb 04

Well, it took me some time to understand how they reached the 2^n - n -1 number.
There is obviously a reference to the set of the parts of a set. That is the set of all possible subsets. And this grows as 2^n. As Craig points out, in a social setting we might not be interested in ALL subsets that can be generated, but only in some. In particular, 2^n also consider the empty subset. Is this socially important. Well some people might say so, but if we don't consider it in our calculation we move from 2^N to (2^N)-1. Also we could ignore the sets generated by only 1 person. Again, it is a personal decision, if they should be counted or not. But since the context is groups that are generated using this software, they probably should be ignored (unless the software has crashed at the time). So under those assumption we have to eliminate N subgroups too.
This moves us to (2^N) - N - 1. That is the number said in the Pdf file. I still think that when he refers wo the "rate of growth" he should speak about either the first derivative of the function:
or the difference between the values with N element and N-1 elements:

Said this, I am sorry you considered crazy the idea to meet me in Brighton.

Posted by: Pietro on 4 Feb 04



MESSAGE (optional):

Search Worldchanging

Worldchanging Newsletter Get good news for a change —
Click here to sign up!


Website Design by Eben Design | Logo Design by Egg Hosting | Hosted by Amazon AWS | Problems with the site? Send email to tech /at/
Architecture for Humanity - all rights reserved except where otherwise indicated.

Find_us_on_facebook_badge.gif twitter-logo.jpg