Feedlack detects missing feedback in web applications

  • Authors:
  • Andrew J. Ko;Xing Zhang

  • Affiliations:
  • University of Washington, Seattle, Washington, USA;University of Washington, Seattle, Washington, USA

  • Venue:
  • Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
  • Year:
  • 2011

Quantified Score

Hi-index 0.01

Visualization

Abstract

While usability methods such as user studies and inspections can reveal a wide range of problems, they do so for only a subset of an application's features and states. We present FeedLack, a tool that explores the full range of web applications' behaviors for one class of usability problems, namely that of missing feedback. It does this by enumerating control flow paths originating from user input, identifying paths that lack output-affecting code. FeedLack was applied to 330 applications; of the 129 that contained input handlers and did not contain syntax errors, 115 were successfully analyzed, resulting in 647 warnings. Of these 36% were missing crucial feedback; 34% were executable and missing feedback, but followed conventions that made feedback inessential; 18% were scenarios that did produce feedback; 12% could not be executed. We end with a discussion of the viability of FeedLack as a usability testing tool.