by Roberto E. Lopez-Herrejon, Alexander Egyed
Abstract:
Recent years have witnessed a convergence between research in SPL and Model-Driven Engineering (MDE) that leverages the complementary capabilities that both paradigms can offer. A crucial factor for the success of MDE is the availability of effective support for detecting and fixing inconsistencies among model elements. The importance of such support is attested by the extensive literature devoted to the topic. However, when coupled with variability, the research focus has been devoted to inconsistency detection, while leaving the important issue of fixing the inconsistency largely unaddressed. In this research-in-progress paper, we explore one of the issues that variability raises for inconsistency fixing. Namely, in which features to locate the fixes. We compute what is the minimal number of fixes and use it as a baseline to compare fixes obtained with a heuristic based on feature model analysis and random approaches. Our work highlights the pros and cons of both approaches and suggests how they could be addressed.
Reference:
Towards fixing inconsistencies in models with variability. (Roberto E. Lopez-Herrejon, Alexander Egyed), 2012.
Bibtex Entry:
@Workshop{DBLP:conf/vamos/Lopez-HerrejonE12,
author = {Roberto E. Lopez-Herrejon and Alexander Egyed},
booktitle = {5th International Workshop on Variability Modelling of Software-Intensive Systems (VAMOS), Leipzig, Germany},
title = {Towards fixing inconsistencies in models with variability.},
year = {2012},
abstract = {Recent years have witnessed a convergence between research in SPL
and Model-Driven Engineering (MDE) that leverages the complementary
capabilities that both paradigms can offer. A crucial factor for
the success of MDE is the availability of effective support for detecting
and fixing inconsistencies among model elements. The importance of
such support is attested by the extensive literature devoted to the
topic. However, when coupled with variability, the research focus
has been devoted to inconsistency detection, while leaving the important
issue of fixing the inconsistency largely unaddressed. In this research-in-progress
paper, we explore one of the issues that variability raises for inconsistency
fixing. Namely, in which features to locate the fixes. We compute
what is the minimal number of fixes and use it as a baseline to compare
fixes obtained with a heuristic based on feature model analysis and
random approaches. Our work highlights the pros and cons of both
approaches and suggests how they could be addressed.},
pages = {93-100},
doi = {10.1145/2110147.2110158},
file = {:Workshops\\VAMOS 2012 - Managing SAT Inconsistencies with HUMUS\\Towards Fixing Inconsistencies in Models With Variability IEEE-preprint.pdf:PDF},
keywords = {FWF P21321, EU IEF 254965},
}