Supplementary Material for Requirements documentation containing natural language: A Systematic Tertiary Literature Review

Context: Requirements documentation in natural language has diverse artifacts, but few studies address their suitability to typesof requirements or ease of communication.

Methods: We conducted a systematic tertiary literature review (STLR) and identified 22 relevant review papers that address natural language artifacts used by practitioners to document software requirements. We also investigated which types of requirements are addressed by artifacts and if there are guidelines for each.

Results: A variety of artifacts used for this purpose were identified, of which the most referenced in the literature were diagrams,use cases, conceptual models, user stories, and prototypes. The analysis highlighted that artifacts are applied differently to functional and non-functional requirements. In general, diagrams, use cases, scenarios, and prototypes can be used for both types of requirements, depending on the content (usability, security, etc.). However, user stories and derived artifacts are more recommended for functional requirements and have limitations for non-functional requirements.

Conclusion: Furthermore, the study explored different guidelines, structures, and formats used in documentation artifacts, reflecting the diversity in requirements documentation practices in software projects.