“Human computation” is the idea of employing “human processing power” for tasks that computers are unable to perform. For example, people can complete parts of machine processes such as image and speech recognition or language analysis that are otherwise difficult or impossible for computers to perform. Among strategies for bringing people into machine process in efficient and economic ways, is the idea of creating games that fun to play yet are designed to to aggregate small cognitive contributions of individuals to complete difficult tasks such as. Google Image Labeler is an example of such a game with the main purpose of accumulating accurate labels for images on the web.
Based on an online ethnographic study of what players and critics say about their interactions with the game and a close reading of arguments made by its designers, this study maps the relationship of the designers’ assumptions about motivation and participation; the form and structure of the game such as organization of content and reward mechanisms; and the individual and collective experiences of people who played the game. I identified five major themes in players’ responses to the game: playing because the game is fun; playing because the game is addictive; playing for a good cause; (not) playing because the game is exploitative; and playing to subvert the game and break the game.
By reducing the community of players to a network, Google Image Labeler fails to realize both the individual and collective potentials of participation in a collective activity such as commitment, responsibility, care and learning to name a few. When terms such as social (as in social computing) is used to refer to online environments such as Google Image Labeler certain aspects or possibilities of the social are excluded. The emphasis is on parts and the whole is assumed to be nothing more than the sum of interchangeable parts. Participation is then reduced to “being a part of” or in the context of technology “to be connected to the network”. This characteristic is reminiscent of assembly line in its treatment of individuals as part of industrial machines. The difference is that computational machines do not require the physical labor of the individual but her intellectual labor in order to secure efficiency. The individual is, nonetheless, treated as an interchangeable part in the operations of the machine. The work is similarly meaningless and does not support self-expression, communication, or social interaction.
[1] Nassim Jafarinaimi, “Exploring the Character of Participation in Social Media: The Case of Google Image Labeler,” in Proceedings of the 2012 iConference, (Toronto, Canada: ACM 2012), 72 –79.