Tweet
Recently it has just came to me that the behavior of forum participation can be modeled using methods familiar in microeconomics. I am wondering if there has been any work done in the field, and if not I have a outline of a model that still need a ton of work to generate predictions.
---------------------
Assume a discussion based forum as opposed to social networking location.
The rational forum visitor:
The utility of each visit of forums is the ratio of reading new interesting content divided by the amount of time spend.
The frequency of a visitor to a forum is a positive slope function of rate of posting and new threads, with some hard minimum/maximums. The decision of a visitor to visit a forum at all is dependent on the signal to noise ratio of the forum.
The rational forum poster:
The utility from posting lies from the ability to direct new conversation (thus improving net interesting content) and the pleasure of self expression and perhaps a harder to define measure to convert other posters.
The chance of posting for a visitor is a unique function with the following inputs: Already posted forum ideas, amount of traffic, and perhaps a characteristic function of particular forum cultures (level of aggression, attention span...and so on) that can be assumed to be largely invariant with respect to traffic.
------------------------
Problems this model seeks to address when complete:
1. The utility of subforums, and when should subforums be closed. (eg. below critical values of activity and people just forget them)
2. What are the natural evolution of forums that is universal amongst its type.
3. The optimized posting pattern to maximize utility of a individual or collectively and by extension the rules and moderation enforcement required.
4. Whether the construction of roboliberal would improve the forums, and many, many other things.
-------------
So, is there hope in this project, even if it is just opening up a new path for KH to call everyone a "****" for not following logical posting patterns? Ideas?
---------------------
Assume a discussion based forum as opposed to social networking location.
The rational forum visitor:
The utility of each visit of forums is the ratio of reading new interesting content divided by the amount of time spend.
The frequency of a visitor to a forum is a positive slope function of rate of posting and new threads, with some hard minimum/maximums. The decision of a visitor to visit a forum at all is dependent on the signal to noise ratio of the forum.
The rational forum poster:
The utility from posting lies from the ability to direct new conversation (thus improving net interesting content) and the pleasure of self expression and perhaps a harder to define measure to convert other posters.
The chance of posting for a visitor is a unique function with the following inputs: Already posted forum ideas, amount of traffic, and perhaps a characteristic function of particular forum cultures (level of aggression, attention span...and so on) that can be assumed to be largely invariant with respect to traffic.
------------------------
Problems this model seeks to address when complete:
1. The utility of subforums, and when should subforums be closed. (eg. below critical values of activity and people just forget them)
2. What are the natural evolution of forums that is universal amongst its type.
3. The optimized posting pattern to maximize utility of a individual or collectively and by extension the rules and moderation enforcement required.
4. Whether the construction of roboliberal would improve the forums, and many, many other things.
-------------
So, is there hope in this project, even if it is just opening up a new path for KH to call everyone a "****" for not following logical posting patterns? Ideas?
Comment