Design Decisions by Voice: The Next Step of Software Architecture Knowledge Management
Paper in proceeding, 2020

© 2020, IFIP International Federation for Information Processing. Architectural Design Decisions (ADDs) capture the essence of relevant Architectural Knowledge (AK) and the underpinning rationale in order to produce well-designed software architectures. AK and design rationale might get lost if not captured at the same time when the architecture is discussed and modeled in early design phases. For years, this relevant knowledge has been captured using text templates and supported by a number of research tools. Nevertheless, as no commercial tool is still available combining AK capturing with UML notations to facilitate capturing the design decisions at the same time the architecture is modeled, is the major barrier to convince software architects and companies to invest in documenting the significant design decisions. As capturing AK using text templates requires an extra effort, we propose an approach to make the documentation process easier and reduce the effort thereof by using voice commands. In particular, we suggest an approach to: (i) capture ADDs using voice commands during design conversations, and (ii) link the captured ADDs to UML notations. Our approach integrates OctoUML, a modeling tool with voice commands for capturing design decisions by voice.

Design Decisions

Software architecture

Knowledge capturing

UML

Architectural knowledge

Voice decisions

Author

Rafael Capilla

Chalmers, Computer Science and Engineering (Chalmers), Software Engineering (Chalmers)

Rey Juan Carlos University (URJC)

Rodi Jolak

University of Gothenburg

Chalmers, Computer Science and Engineering (Chalmers), Software Engineering (Chalmers)

Michel Chaudron

University of Gothenburg

Chalmers, Computer Science and Engineering (Chalmers), Software Engineering (Chalmers)

Carlos Carrillo

Technical University of Madrid

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

03029743 (ISSN) 16113349 (eISSN)

Vol. 12481 LNCS 166-177
9783030642655 (ISBN)

8th IFIP WG 13.2 International Conference on Human-Centered Software Engineering, HCSE 2020
Eindhoven, Netherlands,

Subject Categories (SSIF 2025)

Computer Sciences

Computer Systems

DOI

10.1007/978-3-030-64266-2_10

More information

Latest update

11/19/2025