| 1. |
- Berger, Thorsten, 1981, et al.
(författare)
-
Usage Scenarios for a Common Feature Modeling Language
- 2020
-
Ingår i: SPLC '19: Proceedings of the 23rd International Systems and Software Product Line Conference - Volume B. pp. 174–181. - New York, NY, USA : ACM. - 9781450366687
-
Konferensbidrag (refereegranskat)abstract
- Feature models are recognized as a de facto standard for variability modeling. Presented almost three decades ago, dozens of different variations and extensions to the original feature-modeling notation have been proposed, together with hundreds of variability management techniques building upon feature models. Unfortunately, despite several attempts to establish a unified language, there is still no emerging consensus on a feature-modeling language that is both intuitive and simple, but also expressive enough to cover a range of important usage scenarios. There is not even a documented and commonly agreed set of such scenarios. Following an initiative among product-line engineering researchers in September 2018, we present 14 usage scenarios together with examples and requirements detailing each scenario. The scenario descriptions are the result of a systematic process, where members of the initiative authored original descriptions, which received feedback via a survey, and which we then refined and extended based on the survey results, reviewers' comments, and our own expertise. We also report the relevance of supporting each usage scenario for the language, as perceived by the initiative's members, prioritizing each scenario. We present a roadmap to build and implement a first version of the envisaged common language.
|
|
| 2. |
- Thum, T., et al.
(författare)
-
Towards Efficient Analysis of Variation in Time and Space
- 2020
-
Ingår i: SPLC '19: Proceedings of the 23rd International Systems and Software Product Line Conference - Volume B. pp. 57–64. - New York, NY, USA : ACM. - 9781450366687
-
Konferensbidrag (refereegranskat)abstract
- Variation is central to today's software development. There are two fundamental dimensions to variation: Variation in time refers to the fact that software exists in numerous revisions that typically replace each other (i.e., a newer version supersedes an older one). Variation in space refers to differences among variants that are designed to coexist in parallel. There are numerous analyses to cope with variation in space (i.e., product-line analyses) and others that cope with variation in time (i.e., regression analyses). The goal of this work is to discuss to which extent product-line analyses can be applied to revisions and, conversely, where regression analyses can be applied to variants. In addition, we discuss challenges related to the combination of product-line and regression analyses. The overall goal is to increase the efficiency of analyses by exploiting the inherent commonality between variants and revisions.
|
|
