Front-end interface and user-system interaction are factors that must be carefully considered in software development due to their influence in quality of use. On some occasions, it is the first concern addressed by developers, as it comes naturally from the requirements analysis performed with stakeholders. IFML is a standard language of OMG that supports the abstract description of these front-end interfaces, for software applications on different devices. IFML has been used in the context of Model-Driven Development (MDD) and Model-Driven Architecture (MDA) to describe the elements and behavior of interfaces, aiming to generate code for those interfaces. However, it is necessary to investigate the use of IFML in traditional software development, in order to better understand how it is used for modeling front-end interfaces. This article presents an empirical study that aimed to verify the quality of IFML models created based on a subset of requirements of software two web applications. The quality was defined in terms of models' correctness and completeness. The results showed that the correctness of the models was low, varying from 51% to 55%, while the completeness varied from 66% to 69%. In order to better understand the results, we analyzed syntactic and semantic defects found.

No abstract