What happens when you allow the members of the crowd to interact? On this blog we've looked at examples in which each member of the crowd is unaware of the other members participating in the same task, and yet collectively their efforts are aggregated in interesting ways.
Given my previous work with collective viewing patterns on YouTube, which investigated word-of-mouth interactions amongst the consumers of video content, I was excited when I came across a great example of interactions between producers of video content in this great video.
What's great about this video is that much of it could have been produced in a completely crowdsourced way by specifying rules that each participant could follow without ever interacting with the other players (cover camera for 3 seconds, rotate in chair for 10 seconds, look up for 1 second, etc). However at some point there is an actual interaction between members of the crowd when they start showing up in eachother's videos.
Makes me wonder how large something like this could be scaled up without allowing actual interactions?