Search

This paper addresses the possibility of measuring perceived usability in an absolute way. It studies the impact of the nature of the tasks performed in perceived software usability evaluation, using for this purpose the subjective evaluation of an application’s performance via the Software Usability Measurement Inventory (SUMI). The paper reports on the post-hoc analysis of data from a productivity study for testing the effect of changes in the graphical user interface (GUI) of a market leading drafting application. Even though one would expect similar evaluations of an application’s usability for same releases, the analysis reveals that the output of this subjective appreciation is context sensitive and therefore mediated by the research design. Our study unmasked a significant interaction between the nature of the tasks used for the usability evaluation and how users evaluate the performance of this application. This interaction challenges the concept of absolute benchmarking in subjective usability evaluation, as some software evaluation methods aspire to provide, since subjective measurement of software quality will be affected most likely by the nature of the testing materials used for the evaluation. [ABSTRACT FROM AUTHOR]

Some of the concerns people have about A1 are: its misuses, effect on unemployment, and its potential for dehurnanising. Contrary to what most people believe and fear, GIcan lead to respect for the enormous power and complexity of the human mind. It is potentially very dangerous for users in the public domain to impute much more inferential power to computer systems, which look common-sensical, than they actually have. No matter how impressive A1 programs may be, we must be aware of their limitations and should not abrogate human responsibility to such programs. [ABSTRACT FROM AUTHOR]