Collins-Thompson makes the case for slow search engines
Through the use of search engines like Google and Bing, Web searchers have grown accustomed to receiving answers to their questions within a fraction of a second.
However, in his article “Slow Search,” which appears in the August 2014 issue of Communications of the ACM, UMSI Associate Professor of Information Kevyn Collins-Thompson suggests that slow search techniques in many cases can improve the quality and relevance of results.
By utilizing techniques like complex query processing to identify key concepts and derive multiple queries from the initial one, slow search engines can make use of additional response time to employ resources that are inherently slow, such as crowd-based ranking methods that use human judgments, in order to identify the most relevant existing content.
Those searching for a specific website or straightforward fact would still want immediate results to their queries, but people who invest minutes, hours, or even days in more complex or exploratory searches, like planning a vacation or researching a medical diagnosis, may be willing to wait for better results or insights.
For these more expansive searches, slow search results can include more in-depth background material to better understand a topic, or provide the context necessary to resume an ongoing task instead of merely returning a list of links.
Slow search approaches can also be valuable for people with intermittent, slow, or expensive network connections. In such cases, it can be difficult for searchers to employ traditional search strategies, such as rapidly reformulating queries to generate desired results.
By changing the way people experience search, including how they express what they are looking for and how they interact with the information they find, slow search approaches could help searchers spend less time on low-level searching processes and focus more on task completion.
Collins-Thompson co-authored “Slow Search” with Jaime Teevan, Ryen W. White and Susan Dumais of Microsoft Research.