RAA stands for: Research Article Analysis
1. Purpose of the research:
Proposes a method to use interaction patterns in heuristic evaluation, in order to facilitate the process of heuristic evaluation as well as improve the output of heuristic evaluation in terms of providing redesign advice.
The authors overviewed the use of heuristic evaluation as usability inspection methods and the prevalence of using design patterns in interface design respectively, and came up with the idea of mapping Nielsen’s heuristics with subsets of design patterns from Welie’s library. This approach seeks to find the correspondence between each heuristic and one or more design patterns, as showed below. The paper also claimed after several evaluation cycles, a refined correlation will be gained.
3. Main Findings:
A case study of heuristic evaluation of a university website using the proposed method shows it is easier to find direct solutions for usability problems found by evaluators.
After I re-read Nielsen’s procedure of conducting heuristic evaluation, I figured out that this paper is trying to fill the gap Nielsen mentioned as “Heuristic evaluation does not provide a systematic way to generate fixes to the usability problems or a way to assess the probable quality of any redesigns”. I appreciate the authors effort in mapping Nielsen’s heuristics with design patterns, trying to offer a systematic solution to the problem, but I think it a little bit forced to combine them in this way.
First of all, I think seeking redesign solutions through the successful examples is a common sense, and the use of pattern library as a reference could be a good idea (which might have already been widely used). But I am not sure how practical and efficient to use the mapping methods proposed in this paper as an evaluation step. After all, Nielsen’s heuristics is subjective, Welie’s patterns categories are subjective, and this mapping between them is subjective, it might be hard and unnatural to force people understanding and accepting the mapping. At least, as I looked at the mapping figure, I could not tell directly and clearly the relationship between the left column and right column. With a specific usability problem found, I would say it is more straight forward to find corresponding examples directly in the pattern library.
Anyway, I will try to test this idea when I do my heuristic evaluation, with questions like this: Is the mapping a redundant step? Will it be easier to locate solutions directly in the pattern library without mapping to their categories first?