Photo: John Koller, KAI Consulting
Technology should provide tools that facilitate the power of the human mind in ways that stimulate creativity in ways that emphasize visual information as a form of thought, through looking, seeing, understanding. That was the central focus of the SCAR 2006 opening address in Austin, Texas, today by recognized expert in user interface design and evaluation Ben Shneiderman, PhD, professor, Department of Computer Services, University of Maryland Founding Director, Human-Computer Interactive Laboratory (HCIL).
Shneiderman said that belief is contrary to the common view on technology which holds that if technology is extremely intelligent, it can help us. He scoffed at this notion stating that “machines are no more intelligent than pencils.” Rather, he said, by empowering users instead of emphasizing computers, we can “enable people to do things a thousand times better than ever before.” Creativity Support Tools must be developed with creativity in mind, not just productivity, he said, adding that it’s not A.I. (Artificial Intelligence) that matters but rather U.I. (you and I) that are truly important.
Through his research group, countless studies have been performed — in an interdisciplinary manner — to better understand how technologies are used. The big push these days, he said, is on mobile devices and web-based applications. Overall, his work and conclusions are heavily influenced by the revolutionary work of Leonardo who held the belief that “the eye is the best way to understand the world.” Therefore, information visualization supported by dynamic technologies is the name of the game.
He provided multiple examples of applications that are dynamically capable of providing information display that, if done properly, can spur better thinking. The first is an application called Spotfire which gives users a type of ‘scattergram’ of information that is able to give users rapid, incremental, and reversible visualization of data sets. The basic premise behind these types of examples (which are not that new, Spotfire has been around for 10 years) is to give the following capabilities: data overview, zoom and filter, and finally details on demand, along with export and import capabilities. An example of a “Visual Analytics” application his group has used is available at www.cs.umd.edu/hcil/hcc/
Other examples of dynamic data visualization applications Shneiderman provided were:
- Tree Maps – color-coded interfaces for better understanding trends, clusters and data outliers with broad query capabilities;
- Temporal Maps – A type of visual representation based on events or time makers. This type of application could be used to show a particular event in patient care with detailed information regarding a problem and how it was addressed, by whom, and what care was administered; and
- Other applications have the potential to work with images that could be used to represent many patients. Shneiderman believes these types of technologies could ease the creation of annotations for more complete medical records.
In general, some of the desirable available tools of today according to Shneiderman include: abundant displays, multiple coordinated windows, dynamic queries, easy data export and import, and powerful annotation and collaboration. Looking ahead next-generation advances should emphasize incorporation with statistical tools and data mining, they should accommodate diverse data sets, and ideally placed in an integrated fashion into organizational workflow, he said.
Technology should encourage impulse and exploration were a recurring theme in the examples he provided. Google, he said, is a good example of a widely used tool that obviously places users in a situation where exploration is made easy. Interfaces for medical informatics could be established to work in similar ways and that is one of Shneidermans’goals.
Because these types of sophisticated tools exist, he said that he has put forward the idea that a World Wide Med should be developed as opposed to a World Wide Web. His vision is that all patient medical data should be accessible globally much like the airline industry, regardless of airline or language, is able to pull up any ticket reservation in 15 seconds. The healthcare industry has largely stalled out in developing interconnected electronic records — with a few exceptions — but what is needed are coordinated interfaces, data standards,