Vicarious reputation growing from a virtual economy for reputation system points?

It's an interesting idea that Josh Ledgard learned is being used in China, and blogged about here.   I like that it tries to make the reputation system points more useful/valuable, and encourages people to ask questions they really want answered -- that is, it might help keep the "signal to noise" ratio strong. I do wonder: would people ever view the spending of points as a disincentive to asking questions?

Another thought: if there were a mechanism to show how many/how often a person got points from others (and who they came from), this might create an interesting "friend of a friend" information set. Basically, it could be a social networking element that would help people assign a sort of vicarious reputation layer to the pure numeric reputation score. 

For example, imagine "Bob" was broadly influential in an online community.  He might get points from lots of other people for answering questions, especially including questions with reward points.  Being able to see all the people (perhaps visually, in a node map?) from whom Bob got his reward points -- and who all those people got their rewards points from, too -- would give you some true sense of the scope of Bob's influence.  He's a legitimate community superstar. 

On the other hand, imagine "Erasamus" had a decently high reputation score, but got half his points from answering the questions of "Maria", who had a lot of points to give away.  Could be that Erasamus earned those points by answering a really hard question for Maria.  Or, there might be some other reason why Erasamus had a high reputation score -- perhaps he asked and answered a ton of questions, but they were lame answers to a bunch of lame questions from Maria. Or maybe Erasamus and Maria had some kind of side arrangement, where she would give him a large chunk of points for some unknown reason...? 

Whatever the case, you would be able to see that Erasamus lacked the kind of broad influence that Bob had, which might make you more willing to trust Bob yourself.  If this information were available, it might even lead to a self-reinforcing vicarious reputation -- because people with both a high quantity score and also a high quality score might see their higher community trust paying off as more influence.  That is, more people might select Bob's answers as best answers in part because of the perceived quality of his reputation. 

The scenario is just in my head, of course.  But the potential interaction reminds me of Jon Berry and Ed Keller book, The Influentials: One American in Ten Tells the Other Nine How to Vote, Where to Eat, and What to Buy.

By the way, a great resource for learning about reputation systems is the University of Michigan's Reputation Research Network hosted by the School of Information.  Lot's of substantive material online, including academic papers and a membership directory with links to homepages/other blogs to help you connect in the reputation community!