Harnessing collective intelligence in Web 2.0

Collective intelligence is the idea that a large group of people, working independently or together, can produce knowledge, solutions and decisions that are better than those of a few experts. Tim O'Reilly listed "harnessing collective intelligence" as a core principle of Web 2.0: the Web gets better the more people use and contribute to it.

How Web 2.0 harnesses it

• Wikis (Wikipedia): Thousands of volunteers write and correct articles; mistakes are fixed by the crowd.
• User reviews and ratings: Amazon and Daraz reviews, IMDb ratings and TripAdvisor help others decide.
• Tagging and folksonomies: Users tag photos and links (Flickr, Delicious), creating a shared classification of content.
• Social bookmarking and voting: Digg and Reddit rank content by votes.
• Blogs and comments: Opinions, discussions and news from many people.
• Link-based ranking: Google's PageRank uses links made by millions of site owners as votes of importance.
• Implicit contributions: Recommendations ("customers who bought this also bought...") and spam filters learn from the behaviour of all users.
• Crowdsourcing and open source: OpenStreetMap, Linux and Stack Overflow rely on contributions from volunteers.
• Q&A sites and prediction markets: Quora and Yahoo Answers pool answers from many users; prediction markets combine many people's bets to forecast events.

Many users --> contribute (edit, tag, rate, link)
          --> system aggregates the data
          --> better content for all users (network effect)

Benefits and issues

Benefits: low cost, wide knowledge, fast updates, and better services as the user base grows (network effect). Issues: vandalism, spam, bias, and false information, which require moderation and reputation systems.