[Air-l] info flow basements and ceilings

Ronald E. Rice rrice at scils.rutgers.edu
Sun May 26 17:55:52 PDT 2002


> Message: 3
> From: "Ben Davidson" <bendavidson at totalise.co.uk>
> To: <air-l at aoir.org>
> Subject: Re: [Air-l] feedback, overload, and equilibrium in online nets
> Date: Fri, 24 May 2002 18:02:09 +0100
> Reply-To: air-l at aoir.org
>
> Ronald,
>
> Is there some rule of thumb that emerged from your work regarding basement
and ceiling levels?

I didn't parameterize the basement and ceiling levels, but did parameterize
the inflows, outflows, within group flows, and a four-fold typology derived
from that: isolates, receivers, transmitters, and carriers.  Isolate groups
received and sent less than average (determined by the parameters of a
log-linear model requiring reciprocity between groups), receives received
more but sent less, transmitters sent more but received less, and carriers
sent and received more than average.  It turned out that in order to sustain
a "carrier" mode, you have to continue sending, even early on when you don't
get reciprocal messages.  Only if you start out a transmitter, of the other
three categories, do you have any chance of becoming a carrier.
Essentially, I found that, as with material settings, the system is
entropic: you have to contribute, on the hopes you will get sufficient
returns, if you want to become a central actor (carrier).  Remaining a
receiver quickly leads to becoming an isolate.  In the information
environment, you really need to be a "pay it forward", or "collective
benefit" kind of participant to maintain a central role.  There are other
articles about online critical mass, and collective benefits, such as
Markus, and Rafaeli and LaRose (who show what factors influence the
longevity of bulletin boards, which require altruistic contributions but can
suffer from tragedy of the commons), whose references I can provide.  I can
send you a print version of the article if you (or anyone else) wants.





More information about the Air-l mailing list