Research into Interactive Systems and their Design

During the last ten years I’ve been fortunate to have opportunities to study a wide range of aspects of interactive systems design and HCI. Here are some projects that I’ve found particularly rewarding:

Camera-based scanning

I became interested in using video cameras for document scanning in the early 1990s, and worked with several colleagues (notably Mike Taylor, Chris Dance and Stuart Taylor) to develop a tool for capturing document images off the desk, enhancing them and recognizing the text, so that it could be pasted into an electronic document. A resulting product, PageCam, was launched in 2000 and is shown below.

page cam photo

Learning from engineering research

In 1991 I came across Walter Vincenti’s fascinating book on aeronautical engineering research, What Engineers Know and How They Know It. I was struck by the differences I saw with interactive systems research. I began an investigation of these differences, and how engineering research methods might strengthen HCI. This has led into work on critical parameters and on improving the review process, both summarized below. Currently I’m working with Walter on a short paper about this successful application of the history of one field of technology to current research in another.

Walter Vicenti\'s Book - What Engineers Know and How They Know It

Critical parameters of interactive applications

How can we make sure that each new release of a software product is a real improvement on the existing release? The answer is, we must discover what to measure. In engineering domains, the critical parameters for measuring performance are well-known, but for interactive systems this is hardly ever the case. In 1996 I began a project to identify the critical parameters of applications and use these as a basis for design. In the design of PageCam, for example, we aimed to reduce the time spent copying text from a paper document; we developed and published an auto-completion method that out-performed the re-typing of phrases 4 or more words in length (see below).

Critical parameters of interactive applications

Studying and improving the review process

In my spare time, I’ve become involved with ACM SIGCHI in trying to improve the process by which papers are reviewed for the annual CHI Conference, the major HCI event of the year. In a 1994 study I compared the contributions reported in CHI papers with those reported in engineering research papers, and found dramatic differences, most CHI papers reporting designs and heuristics while most engineering papers reported improved analytical models. Ten years later this is still the case, but meanwhile CHI has adopted a contribution-based approach to reviewing, pioneered by me and Kate Ehrlich when we were co-chairs of the CHI ’99 papers program. This year I set up a SIGCHI subcommittee, the CHI Papers Support Team, to help introduce and evaluate further improvements to the review process.

Studying and improving the review process

Busy Days studies

These are reported elsewhere on this site.

Posted November 20th, 2005 + plink