WebDB - Related public Items

TC Administrator

This item is part of the public WebDB named TC Administrator. The public items below all share the keyword science.

6795 - Harnessing The Wisdom of Crowds in TeleCommunities

07/02/2005 - 20:23:44

James Suroweiki's book The Wisdom of Crowds outlines the concept that "large, diverse groups of people are smarter than an elite few, now matter how brilliant that elite few may be. The are better at solving problems, spurring and fostering innovation, and even predicting the future" (looslely quote from blurb in dust jacket of The Wisdom of Crowds)

I want to explore some possible future experimental uses for telecommunities based on these, andoter concepts here in this posting.

Suroweiki in The Wisdom of Crowds explores evidence that shows that the best way to harness the "wisdom of crowds" is to make the knowledge of individuals useful on a larger scale, yet to still allow the knowledge and information to remain "resolutely specific and local".

Each individual in a Telecommunity collects info on an individual scale, with local expertise and knowledge unique to themselves and their local life conditions. The "wisdom of crowds" can be applied bets by figuring out ways to aggregate that knowledge both manually, and automatically through functions in the software or a combination of both.

There are important concepts to consider in applying this collective wisdom concept to telecomminities:

Individuals should be diverse. Or, more often realistically, a group of individuals should be large enough to ensure diversity.

Individuals should work seperately toward the same goal. Example: all of the people involved should research an issue or question and come to their own conclusions.

These conclusions should then be aggregated based on a set of basic criteria, preferably by a seperate group of diverse individuals from the group that made the initial conclusions. This actually gives different local telecommunities, and different individuals from within the larger telecommunity network ongoing opportunities to work on different projects within the larger network.

The system itself wouldn't necassarily be totally decentralized, as there would be a group of people who would initiate the process each time. Perhaps a city is looking for the best data on where to spend a given amount of money? One individual may then approach a telecomminity that he is a member of. That Telecommunity may then collectively create a research proposal for the city, including total costs of research, etc. They could do this using the "wisdom of crowds" concept, by inviting everyone in the telecommunity (and anyone else they want) to create proposals, then aggregating those into one proposal that they present to the city. They may invite other individuals or telecomminties into the process at their discretion. The individuals creating their own proposals would be could use Telecommunity software to post their proposals. (Perhaps the could purchase the right to use pre-existing knowledge products within the Telecommunity network database knowledge market, for example?) Then, they can create an aggreageted complete proposal colecting the best wisdom and insights of each proposal to make a proposal bid for the city to research how the City's money can best be spent.

If the City approves the proposal, and pays to fund the research, then the telecommunity can then create an open and paying project that allows people from both within and from outside of the city to research and conclude how the money could best be spent. They might also use some type of online poll that polls local citizens of the city in question for their input, etc. Ultimately, the best content of all research is aggregated into one main research analysis and conclusion by the telecommunity, and given to the city. If the research was executed by a diverse group working independently, whose best insifghts were aggregated by another diverse group, the finsished product should be close to as comprehenisive as possible.

Telecommunities,a nd TeleCommunity software provides an environment that is very conducive for this type of process, particularly the "scaling" of the project. Telecommunity software may need to be developed a bit more to harness some of these concepts, btu I believe it is not far off.

Another example is illusttrated by Dave Pollard in his blog. He starts with the frame work that NASA uses for decision making, and describes areas where "wisdom of crowd" concepts can be inserted into the process. (Thanks to Howard Rheingold for pointing out the Pollard article to me.)