Unleashing the Power of Implicit Feedback in Software Product Lines: Benefits ahead
Software Product Lines (SPLs) enable the development of an entire family of software products through systematic reuse. `Reuse’ encompasses not only code but also the transfer of knowledge acquired in one product to others within the SPL. This transfer includes bug fixing, which, when encountered in one product, impacts the entire portfolio of the SPL and enhances the overall code quality. Similarly, feedback derived from customers’ usage can provide valuable insights that benefit the entire SPL when collected from a single product.
This paper explores the potential of incorporating implicit feedback derived from user behavior and interactions into SPLs. Implicit feedback refers to the automated collection of data on software usage or execution, which allows for the inference of user preferences and trends. While implicit feedback is commonly utilized in single-product development, its application in SPLs has been overlooked. We argue that feedback needs to be captured at the platform level and then integrated into the code of individual products during product derivation. This approach aligns with the platform-centric nature of SPLs, where product configurations determine the code rather than direct coding. The paper identifies a set of SPL activities that can benefit the most from implicit feedback. We contrast this hypothesized usefulness with practical evaluations conducted with practitioners using a questionnaire-based assessment (n=8). The results provide positive insights into the advantages and practical implications of adopting a “platform-centric feedback” approach to SPLs and reveal that privacy concerns are a significant consideration for practitioners.
Mon 23 OctDisplayed time zone: Lisbon change
16:00 - 17:30 | |||
16:00 30mTalk | Virtual Domain Specific Languages via Embedded Projectional Editing GPCE | ||
16:30 30mTalk | Unleashing the Power of Implicit Feedback in Software Product Lines: Benefits ahead GPCE Raul Medeiros University of the Basque Country (UPV/EHU), Oscar Diaz University of the Basque Country, Spain, David Benavides Universidad de Sevilla | ||
17:00 30mTalk | Automatically Generated Supernodes for AST Interpreters Improve Virtual-machine Performance GPCE Matteo Basso Università della Svizzera italiana (USI), Switzerland, Daniele Bonetta Oracle Labs, Walter Binder USI Lugano |