{"id":306218,"date":"2010-04-12T06:00:20","date_gmt":"2010-04-12T13:00:20","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?p=306218"},"modified":"2016-10-16T18:32:42","modified_gmt":"2016-10-17T01:32:42","slug":"chi-2010-highlights-hci-diversity","status":"publish","type":"post","link":"https:\/\/www.microsoft.com\/en-us\/research\/blog\/chi-2010-highlights-hci-diversity\/","title":{"rendered":"CHI 2010 Highlights HCI Diversity"},"content":{"rendered":"

By Janie Chang, Writer, Microsoft Research<\/em><\/p>\n

One glance at the list of topics featured in the CHI 2010 Technical Program (opens in new tab)<\/span><\/a> is all it takes to understand the diversity of research disciplines that contribute to the field of human-computer interaction (HCI). CHI 2010 (opens in new tab)<\/span><\/a>, the Association for Computing Machinery\u2019s Conference on Human Factors in Computing Systems, brings top HCI researchers to Atlanta from April 10 to 15 for six busy days of presentations, panels, exhibits, and discussions.<\/p>\n

CHI challenges participants to advance HCI research with work that supports the diversity of human experiences in computing and empowers people from all walks of life. As a result, the technical papers featured during the conference represent a fascinating mix of approaches that range from new input\/output methods, to user behaviors, to applications for specific environments. With nearly 1,000 submissions, review committees had the difficult task of accepting just 365 for presentation. The number of submissions is an indication not only of lively interest in this field, but also the variety of issues that form this area of study.<\/p>\n

Microsoft has been involved actively in CHI conferences, both as sponsor and contributor. This year, 38 technical papers submitted by Microsoft Research were accepted by the conference, representing 10 percent of the papers accepted. Three of the Microsoft Research papers, covering vastly different topics, won Best Paper awards, and seven others received Best Paper nominations.<\/p>\n

Best Papers Highlight HCI Diversity<\/h2>\n

A Longitudinal Study of How Highlighting Web Content Change Affects People’s Web Interactions<\/em><\/a>,<\/em> by Jaime Teevan<\/a>, Susan T. Dumais<\/a>, and Daniel J. Liebling<\/a> of Microsoft Research Redmond<\/a>, explores a dynamic approach to viewing Web pages. Teevan also shared the results of this study in a talk at Stanford University (opens in new tab)<\/span><\/a> in December 2009.<\/p>\n

Digital information, especially Web pages, changes quickly and regularly. As a general rule, every year the Web doubles in size and half the existing Web content changes. At the same time, capturing data about change in digital-information artifacts is easy compared with traditional formats, yet existing tools do little to capture, store, or make use of dynamic history.<\/p>\n

\u201cWeb browsers show Web pages only as they currently appear,\u201d Teevan explains. \u201cAnd search engines only index pages as seen at the time they were crawled. We started studying Web-content change to learn more about how we could best take advantage of changing information.\u201d<\/p>\n

The research team formed a study involving 30 participants and a browser plug-in called DiffIE<\/a>, an internal tool not publically available. Developed during one of Teevan\u2019s earlier projects, DiffIE caches pages visited by the user, then, when the user returns to a previously viewed page, highlights changes that have occurred. While there have been other tools designed to track changes to Web sites, this was the first to highlight changes so that the user is constantly aware of new content during normal Web interactions. DiffIE made it possible for Teevan and her team to conduct a study to determine how user interactions with dynamic content might change through consistent awareness of Web-content change.<\/p>\n

In the study, the researchers asked participants to answer the same set of questions before, and one month after, installing DiffIE. The survey questions were designed to elicit information about:<\/p>\n