- |General Program Info
- |Web History |
- |Plenary Speakers
- |Industry Speakers
- |Developers Track
- |W3C Track |
- |Workshops |
- |Sun May 6
- |Mon May 7
- |Tue May 8
- |Wed May 9
- |Thu May 10
- |Fri May 11
- |Sat May 12 |
- |On-Site Logistics
- |Social Events |
Searching Personal Content
Time: Friday, May 11, 2007 (10:30am-noon)
Moderator: Yoelle Maarek (Google Haifa Research Lab, Israel)
- Susan Dumais (Microsoft Research)
- David Glazer (Google Inc.)
- Andrew Tomkins (Yahoo! Research)
- Chengxiang Zhai (University of Illinois at Urbana Champaign)
It has often been argued that it is much easier to find information on the Web rather than on one's own desktop, email accounts, or even Blog postings. One of the earliest pieces of work in this direction can be traced back to Dumais et al. in their now famous "Stuff I've Seen" paper at SIGIR'2003. Since then, desktop search, as well as built-in search support, in operating systems like in MacOS and Vista, have made (or remade) their apparition. While one could have thought that these solutions would solve, once and for all, the issue of searching personal content, it seems that the issue has shifted with the increasing use of Web services to store and maintain personal content on various hosts. Indeed, users are storing more and more of their favorite content, on remote servers not only for the purpose of sharing but also in order to take advantage of free storage and ease of maintenance. People commonly have several Web mail providers, a Blog hosting service, maybe a Web storage service, etc. As personal content becomes distributed and shared with many, it has become harder to search one's own stuff.
This panel will discuss the different means of searching personal content on the Internet (desktop files, chat sessions, email, my own blog, my hosted documents, my Web history, etc.) and investigate the issues associated with this need. Indeed, personal content in its wide definition contains documents scattered across the Web as well as on the desktop, originating from different sources, in different formats and on different hosts.
The key questions that the panelists will address are:
- What are the characteristics of users' needs and expectations when searching personal content?
- Does a unified or federated model need to be devised to answer these needs?
The answers will probably not be simple (and hopefully controversial enough for a lively discussion!), and cover various issues like information needs, user experience, alternate ranking models, latency, privacy, Web search services and formats, and interoperability.