Very interesting article in Read/Write Web about Facebook planning to use the success of an application as a way of choosing how much “spam” (my word, not theirs!) the application can generate.  The idea is that applications that are generally popular with the readers they contact would be allowed to contact more readers in the future.

It would be very interesting to build models to explore how this idea works in practice.  Two dimensions I suspect are particularly rich are:

1) How much throttling of applications is necessary to keep the amount of “application spam” down to a reasonable level, for a given level of new Facebook applications?
2) If a particular user marks a particular application as valuable (usually implicitly, by using it), how might that translate into a score for that application in the future?  Is there a role for user reputation in this game, as well as application reputation?


Powered by ScribeFire.

Written by

Comments are closed.