Where? - Collaborative Web Search: Who, What, Where, When, and Why - page 46

The WeSearch browser supports automatically dividing a webpage into smaller component
pieces, called clips , by parsing the DOM (Figure 4.22). This allows individual images and paragraphs
to be dragged out of the browser in order to reduce on-screen clutter (by enabling users to close the
original pages) and support sensemaking (by enabling users to identify the most relevant content and
put it within containers (Figure 4.23)). The search-by-example feature analyzes containers' contents
Figure 4.23: Containers provide a place for WeSearch users to organize their clips, in order to facilitate
sensemaking; containers providing several different organization templates (such as lists and grids) are
available. To reduce the need for textentry on multi-touch surfaces, container contents are analyzed for
commonalities, resulting in suggested search terms that appear on the container's bottom border.
for similarities, and it proactively suggests new query terms that might be useful. These suggested
queries can be issued without additional typing, by touching the suggestion region. Groups can
also export an XML record of their WeSearch session, which contains information about every
component clip, including which group member found it and how (i.e., using what keywords), and
what container it was organized into. This meta-data rich exported record, viewable in any Web
browser, enables individual, post-meeting sensemaking and reflection.
Evaluations of WeSearch found that the marquee feature enhanced awareness of group mem-
bers' search strategies, promoting conversation, and that clips and containers were successful in
supporting sensemaking as an integral part of the search process. Text-reuse and report-exporting