Crowds, Communities, and Mixed-Initiative Systems

In this talk, I highlight opportunities for designing new forms of crowd-supported, mixed-initiative systems that tightly integrate crowd work, community process, and intelligent user interfaces to solve complex problems that no machine nor interested party can solve alone.

I focus on two examples. My first example is Mobi, a system that coordinates a crowd to plan custom trip itineraries. By using automatically generated todo items to focus the crowd's attention on what needs work, Mobi illustrates a novel approach for handling human computation tasks that are difficult to decompose. 

My second example is Cobi, a system that engages an entire academic community in planning a conference. Communitysourcing applications collect preferences, constraints, and affinity data from community members, and intelligent session-making and scheduling interfaces combine communitysourced data and constraint-solving to enable organizers to make informed decisions when creating and improving the schedule. I will share findings from recent deployments for planning CHI and CSCW, the two largest conferences in human computer interaction.