Comparison of a textual versus a graphical notation for the maintainability of MDE domain models: an empirical pilot study

Abstract

Background: Models are a useful tool to increase the developer's productivity and satisfaction when performing maintenance tasks. However, in order to maximize these advantages, the right selection of notations must be made. Unfortunately, the software engineering field lacks a body of empirical evidence that supports such selection. A suboptimal decision in this regard may have negative consequences over the maintenance process. Goal: To compare a textual and a graphical notation with respect to the efficiency, effectiveness and satisfaction of software developers while performing analysability and modifiability tasks on two different applications. Method: We have carried out a quasi-experiment with 86 3rd-year students of the Computer Engineering degree at the University of Alicante. Subjects were randomly classified in two groups, and each group performed 20 maintenance tasks with a textual and a graphical notation. We measured and compared the efficiency, effectiveness and satisfaction of subjects assigned to each treatment. Conclusions:The analysed data show that the coverage of analysability and the efficiency of modifiability are affected by the type of notation used, regardless of the application. In both cases, subjects using the textual notation performed significantly better. This study needs to be replicated to extend our conclusions to other subject profiles and application types.

Impact of model notations on the productivity of domain modelling: An empirical study

“…It is important to note how these results contrast with the ones obtained in a previous study on maintainability of MDE models using either textual or graphical notations [29], in which the textual notation improved both the efficiency and the effectiveness of the subjects during certain tasks of analysis and modification of domain models. Such contrasting findings (textual notations being superior for model maintainability tasks and graphical notations being superior for model creation tasks) corroborate the claim made by several authors that A C C E P T E D M A N U S C R I P T the cognitive load of the different languages may impact differently depending on the particular task [2,44].…”

“…The results indicated that the semi-structured natural language resulted in the highest efficiency and effectiveness. Also closely related to the study presented in this paper, in [29] it was empirically assessed how the use of a textual vs. a graphical domain-model notation has an impact on the maintainability of the models, with the textual notation outperforming its graphical counterpart.…”

An Empirical Study on the Comprehensibility of Graphical Security Risk Models Based on Sequence Diagrams

“…Comprehensibility is measured with respect to task scores, while efficiency is measured with respect to average time. However, these measurement types for comprehensibility and efficiency are often used in similar studies [21,13,19]. Furthermore, to prevent bias, all experiment material are the same for both groups with the only difference being the graphical or textual annotations.…”

“…Meliá et al [19] compared graphical and textual notations for the maintainability of model-driven engineering domain models in a pilot study. The study was performed with students as participants, and showed that the participants using textual notation performed better with regard to analyzability coverage and modifiability efficiency.…”

PYFML - A Textual Language for Feature Modeling

“…The propositional logic is most commonly used to represent the semantics of feature model to obtain all valid feature configurations where each feature considered as Boolean variable and each feature configuration must satisfy the propositional formulas corresponds to the feature diagram [18], [19]. The study in [20] compares between a graphical notation and a textual notation with respect of efficiency of analyzability and modifiability, and shows that the use of a textual notation is more efficient than graphical notation and in this work, a textual notation has been chosen first for its important and the alternative graphical notation that facilitates expressing and visualizing of feature model is left for future work.…”