We describe an approach to study the long-term use of GUI applications that supplements a log of low-level events with additional data gathered at the applications' architectural boundaries. We implement a preliminary system based on this approach and apply it to two applications. For the second application, we compare the data collected with our technique to data collected with manual instrumentation. We demonstrate that our technique is easy to apply to new applications and captures information missed by manual instrumentation. This additional information is helpful in answering questions about the use of the application. However, our technique generates large logs and does not yet capture all of the information needed to study the use of applications. We conclude with proposals for rectifying these deficiencies in future systems. © IFIP International Federation for Information Processing 2005.
CITATION STYLE
Hartman, G. S., & Bass, L. (2005). Logging events crossing architectural boundaries. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 3585 LNCS, pp. 823–834). https://doi.org/10.1007/11555261_65
Mendeley helps you to discover research relevant for your work.