Privacy In Ubicomp'2003

Ubicomp communities: privacy as boundary negotiation

Sunday, October 12, 2003

Seattle, Washington, USA

Held as part of UbiComp 2003

Ubiquitous computing conjures visions of big and little brother, and
ever-diminishing privacy. But it also opens up new forms of
communication, collaboration and social relations. Full participation
in communities involves exchange of information, and maintenance of a
visible, public persona. Privacy is often regarded as an imperative in
its own right, but this perspective ignores the countervailing need
for disclosure in social settings. This workshop takes a balancing
perspective: it treats community participation as a goal, and balances
the need for disclosure against the need for privacy. Privacy is not
an abstract desideratum, but a process of negotiation of disclosure
boundaries.  The workshop will explore both social perspectives and
technical approaches to this issue. This workshop builds on last s
ubicomp workshop Socially-informed design of privacy-enhancing
solutions in ubiquitous .

This workshop aims to provide a forum for ubicomp system developers
and researchers, security researchers, and social scientists to
collaboratively explore the future of trust-sensitive and community
tools in ubicomp. Areas of interest to this workshop include (but are
not limited to) the following topics:

1.  Community models and Ubicomp: What are the emerging and
anticipated forms of community supported by ubiquitous computing? What
are the communication modes? How are norms established and maintained?
What design principles can be discovered?

2.  Communities and Privacy: What forms of disclosure and discovery
are appropriate for ubicomp communities? How is disclosure mediated?
What kinds of disclosure boundary are appropriate? How can this be
exposed and supported by technology? How can disclosure boundaries be
negotiated? Are there asymmetries that need to be mitigated? This
topic has both social and technical dimensions.

3.  Communities and Trust: With respect to privacy and disclosure, how
manifest in ubicomp communities? How are reputations, reliabilities,
and risks established, measured, and represented? What forms of
information or other exchange occur in the community? How might
ubicomp systems handle differences in power, access, and expertise
within a community?

Workshop Format

This workshop will last for 1 full day and will be limited to 20
participants (not including the workshop organizers) to enable lively
and productive discussions. Participants will be invited on the basis
of position papers. Such position papers should be no longer than 4
pages excluding references, and they will be selected based on their
originality, technical merit and topical relevance.

The workshop will be organized into panels and breakout
sessions. Depending on the submitted position papers, the workshop
will consist of 3 to 4 panels. Each panel lasts about an hour, and
includes presentation of 5 or 6 position papers that share a similar
topic, followed by organizer-moderated discussions. The morning panels
are devoted to community-oriented ubicomp systems, while the afternoon
panels are devoted to trust issues manifested in those systems. Also
in the afternoon, there will be breakout sessions lasting about 1.5 to
2 hours, followed by reports to a plenary session. In addition, coffee
breaks and lunch will serve as opportunities for informal
discussion. To the extent possible, participants will have lunch
together within short walking distance of the workshop location.

Papers should be submitted to in PDF or MS Word format on or before
August 8, 2003 to . It is
recommended that authors limit their submissions to no more than 6
pages, A4 or letter size. Notification of acceptance will be sent out
by August 22, 2003.

Workshop Organizers

    *      John Canny University of California, Berkeley
    *      Paul Dourish University of California, Irvine
    *      Jens Grossklags University of California, Berkeley
    *      Xiaodong Jiang University of California, Berkeley
    *      Scott Mainwaring  Intel Research


Please direct questions to