A. I don’t know. Because he’s on Ebay??? [looking around for help from other students]

B. It’s an interesting phenomnenon, isn’t it? Why do we trust people online?

A. Well, with some of them you get to know them really well even though you never meet them face to face.

B. True enough. There are people I feel like I have known forever that I’ve never seen face to face. We exchange emails, post comments on each others’ blogs, read each others’ writings, etc. After a while you really do feel like you know them.

A. I guess its the same way you come to trust people in the “real world.”

B. Which is… how? How do you know you can trust someone?

A. [with great pride] Well how do you really *know* anything?

B. I think that question leads us down a different path. [thinking again] At what point do you begin trusting someone?

A. After they earn my trust.

B. Okay. Keep going with that. How does someone “earn your trust”?

A. [pauses] Well, I guess through chances to do me wrong. I mean, if someone has an opportunity to really take advantage of me, and they don’t, then I start to trust them.

B. What’s the difference between that person and another who you might trust with something extremely important to you – say your car, house, or spouse?

A. I guess the person I trust more has had more chances to take advantage of me, but has proven over and over again that he won’t.

B. So trust is somehow the product of a person’s past interactions with you?

A. Something like that.

B. Would knowing the record of a person’s past interactions with *other* people influence your willingness to trust them?

A. You mean like if it turned out they had been stealing from their boss or cheating on their wife or something?

B. Or if they won a Nobel Peace prize, etc.

A. Well, sure. If I’ve never met a person before, then I don’t have that history of interactions with him. I guess finding out how he had interacted with others in the past would be all I had to go on.

B. So knowing a person’s reputation can be a substitute for your own personal history of interactions with them?

A. To some extent. I don’t think I would trust other people’s experience as much as I would trust my own, but it’s better than nothing.

B. Let me recast the discussion a little. This all started with the question “would you be comfortable giving your money to someone?” If we change the question to “would you be comfortable taking somone’s answer to your question as authoritative?” does it change your answer at all?

A. You mean like in the newsgroups, when people would ask questions and then have them answered by total strangers? So the question is why take their word for it?

B. Sure.

A. Well, if they answer, they must know what they’re talking about.

B. [grinning] Think about what you’re saying a bit more.

A. I don’t know. A lot of times you can get a clue from the words they use. To be simple about it, if they use big words like they know what they’re talking about, I guess I think that they probably do.

B. So if a person uses one set of words instead of another (to say *exactly* the same thing) it can come off as more believable?

A. Yea. Actually, I tried this approach in most of the essay questions in history as an undergrad. [general laughter and head nodding around the room]

B. Well, hopefully we can agree that the decision to enter into a financial transaction with a stranger is generally an important one. So is the decision to believe an opinion expressed by a stranger on questions where the accuracy of the answer is important.

A. Granted.

B. So…

A. [cutting in] Ok, I get it! You’re going to say something like [clearing throat, and then proceeding in an unnaturally deep voice] ‘people don’t interact very effectively if they can’t trust each other – always watching their backs, etc. So for social software to really let people interact effectively, the software has to deal with the problem of trust.’

B. [bowing to A.] Herr Professor.

[applause from the rest of the class for A.]

[to everyone] You won’t be surprised to find out that lots of people have thought about this problem.

A. Or to find that they’ve written articles about the topic that we have to read…

B. Thanks for the segueway. Try these articles first:

* Kollock’s The Production of Trust in Online Markets
* Reputation Systems by Resnick et al.
* Trust Among Strangers in Internet Transactions: Empirical Analysis of eBay’s Reputation System, also by Resnick and Zeckhauser

A. So let me guess. Those were the readings for the week, which means that next is…

B. The week’s assignment!

Assignment: Synthesis 1

Write a brief piece describing the relationships between cooperation, incentives, reputations, and trust. Focus on connecting the concepts and frameworks in the articles you have read for this class with your own personal experiences. Questions you could answer INCLUDE BUT ARE NOT LIMITED TO: In what manner are reputations and trust incentives in and of themselves? Are reputations any more important in facilitating cooperation online than offline; why or why not? Et cetera.

Don’t forget to post your short piece on your blog.

A. [sarcastically] Could you possibly ask a broader question, please?

B. That’s part of the point. I want to give you a fairly broad license to show that you’re really doing some thinking. Synthesize. Connect your experiences to what you’ve read.