JJ Jacobi
 |
|
Biography
JJ Jacobi is currently consulting with organizations and
startups that use group and social computing as a core component of their
business. From 1997-2002, JJ was a director of software development at
Amazon.com creating the personalization, social computing and shopping metrics
technologies used by shoppers over Amazon’s ever growing product selection. JJ
developed various approaches to individualized recommendations, customer
reviews, reputation systems, customer-created lists, auctions, and group-based
trends. These projects relied on inventing the ability to run real time
experiments to measure how new services and design treatments changed session
shopping behavior. Before Amazon.com, JJ worked at mFactory and Kaleida Labs
(an Apple & IBM joint venture) delivering cross platform media and game
authoring environments. These products grew her interest in tools that allowed
a wide array of creative talent to join their efforts together into a finished
product. During and after university, JJ was at Sun and SGI researching and
developing new 3D and 2D desktop environments that shipped with Sun and SGI
workstations. During university years JJ received several honors for
contributions made to development of the university’s computer environments.
This is also where her interest in social computing began—watching the
university netnews hub and gathering statistics that differentiated the news
groups that were successful in information and social exchange.
Position Paper
Social computing is about enriching specific activities that
communities of people already participate in. It should increase the
entertainment and efficiency of a behavior by allowing it to be done just as
easily (or more easily) as before, but with more accuracy, information, and
creativity than without the community. Like personalized recommender systems,
it helps bring the “wisdom of the crowd” to your electronic doorstep with the
only difference being the level of anonymity associated with the final result.
Both kinds of services have an influence on the way that
people behave and create their social identity and relationships. This makes
data security, context, and the preservation of time and self-worth important
considerations in their design. In this respect, it is beneficial to require no
or minimal participation in order to benefit from a community or network. The
richness of what is received from the community is what sparks more input into
the system and in return even more rewards. Large upfront contributions of data
or trust introduce a high amount of friction to participate. The result is a
decrease in the growth and frequency of rewards for the members.
Information from social computing networks should be
integrated into the actions we are already interested in; not something that
needs additional checking or requires additional complexity to participate.
Measurement and iteration are requirements to developing an understanding of
how these systems work since the way we act in a new world comes prior to
understanding why we act that way.
Back to Social Computing Symposium 2005
|
