Tuesday, November 16, 2004
Marc Smith
Our class was required to attend to lectures by Marc Smith, a sociologist from Microsoft. One of the lectures given in class focused primarily on game-theory or the prisoner's dilemma. For those who didn't know the concept, he gave an interesting exhibition using two members of the class. To both, he gave them an envelope and dollar and sent them out of the room with two instructions. First, they could choose to keep or put the dollar in the envelope. Secondly, when they came back in the room, they would then trade envelopes, and he would double the money (if there was any) in the envelope. Basically, it's a game of trust. If both people trust each other, they will both put the dollar in the envelope and stand to both earn two dollars. If they both keep the dollar because they don't trust the other, then they will only have one dollar. The final scenario, which is the one that played out in class involves one student keeping the money while the other succumbs to trust and puts the money in the envelope (hoping all the while that the other one has too). Instead, he opens his envelope, finds nothing, and realizes the other person has now gotten away with three dollars (the one she kept, the one in the envelope, and the promised extra dollar). Kind of a lengthy description, but hopefully understandable.
The point in bringing up game-theory was for Smith to elucidate the idea that the Internet contains lots of instances in which game-theory comes into play. Sites such as e-bay rely on the trust of both buyer and seller, and when one fails to live up to that trust, the person who trusted is screwed. Sites like ebay have to work to manage this sort of problem.
Smith also spoke in a lecture outside of class on the topic of computer mediated collective action. Much of his discussion focused on his observations of UseNet postings and the various sociological data that can be gathered from people who use UseNet. He showed many charts and graphs detailing how many times people had posted and how many times they visited UseNet in the past month, whether they replied or answered to other posts, etc...
In all, it was fairly interesting, and made me wonder if being a sociologist was always as interesting as he seemed to make it...
The point in bringing up game-theory was for Smith to elucidate the idea that the Internet contains lots of instances in which game-theory comes into play. Sites such as e-bay rely on the trust of both buyer and seller, and when one fails to live up to that trust, the person who trusted is screwed. Sites like ebay have to work to manage this sort of problem.
Smith also spoke in a lecture outside of class on the topic of computer mediated collective action. Much of his discussion focused on his observations of UseNet postings and the various sociological data that can be gathered from people who use UseNet. He showed many charts and graphs detailing how many times people had posted and how many times they visited UseNet in the past month, whether they replied or answered to other posts, etc...
In all, it was fairly interesting, and made me wonder if being a sociologist was always as interesting as he seemed to make it...
# posted by David Duncan @ 10:48 AM 
