Identifier Name Similarities: An Exploratory Study

Identifier names, which comprise a significant portion of the codebase, are the cornerstone of effective program comprehension. However, research has shown that poorly chosen names can significantly increase cognitive load and hinder collaboration. Even names that appear readable in isolation may lead to misunderstandings in contexts when they closely resemble other names in either structure or functionality. In this exploratory study, we present our preliminary findings on the occurrence of identifier name similarity in software projects through the development of a taxonomy that categorizes different forms of identifier name similarity. We envision our initial taxonomy providing researchers with a platform to analyze and evaluate the impact of identifier name similarity on code comprehension, maintainability, and collaboration among developers, while also allowing for further refinement and expansion of the taxonomy.