Traditional user-centered methods don't really address the design issue, but assume that the screen design somehow pops out of the air. The user-centered methods deal with how to find the usability problems in the design. The designer then has to correct the screens, test again, and so on. This approach corresponds to writing programs without plan and structure, and then test and correct them until they work. This trial and error approach has long been abandoned in software engineering, but what is the solution for the user interface?
 
I will show a systematic design approach we have been using for several years by now. It starts with two traditional descriptions: a data model and high-level use cases (task descriptions). From that you design the screens in a systematic way. The result is better task support than traditional approaches, a system that is easier to understand, and fewer screens in total. As with good programming, the result is not error free, but it has only minor problems that are easy to correct.

 
About the Speaker:
Soren Lauesen is professor at the IT-University of Copenhagen. He has worked in the IT industry for 20 years and has been a professor at Copenhagen Business School for 15 years. He has been co-founder of three educational and two industrial development organizations. He has been a member of the Danish Research Council for Science (4 years) and the Danish Research Council for Technical Sciences (8 years).
 
His industry projects have encompassed compilers, operating systems, process control, temporal databases, and software quality assurance. His research interests include human-computer interaction, requirements specification, object-oriented design, quality assurance,Ý marketing and product development, and interaction between research and industry.Ý He has a broad range of other interests ranging from biology to dancing and foreign cultures.