dialogue with interactive practitioners and for user models and tasks that can be used within test
collection or system-based evaluations. A concern was raised that the core metric of relevance is
being stretched too far and we need additional measures to discuss interactive searching. The
work of Toms and O’Brien was signaled as a way of defining new metrics from careful user
studies. In addition, we should not isolate online searching to search engines alone, but also
include other sources of information with different models, such as those used by Amazon.com.
There was also recognition that the process of searching is as important as the end-product of
searching.
Throughout the workshop the discussion demonstrated that Web Information Seeking and
Interaction is both a complex area and one that requires input from many sources. We are,
therefore, grateful for the input from our panelists, attendees, and contributors to the proceedings
who gave us a lively day of discussion and insights from different parts of the community.