Involved researchers

  • Dr. Pascal Costanza
  • Jorge Vallejos
  • Engineer Bainomugisha
  • Dr. Eddy Truyen
  • Dr. Riccardo Scandariato
  • Frans Sanen
  • Prof. Patrick Heymans
  • Andreas Classen
  • Arnaud Hubaux
  • Dr. Sebastian Gonzalez
  • Alfdredo Cadiz
  • Antoinne Marot
  • Nicolas Maquet
INRIA-Futures (EU partner)
  • Guillaume Wagnier
  • Carlos Parra

Partner research interests

In this section we record the members’ research interests with respect to the topics of software composition and software decomposition

Eddy Truyen:

  • Software composition: In the context of unified aspect-component models, Eddy Truyen studies dynamic reconfiguration of aspects and components in distributed systems. Previously, for his PhD, he has studied context-sensitive composition of aspects and components.
  • Software decomposition: In the context of feature models, Eddy Truyen studies the documentation and management of feature interactions.

Patrick Heymans, Andreas Classen and Arnaud Hubaux:

  • In this WP, we are mostly interested in the topic of separation of concerns (SoC) in the early stages of software development, known as requirements engineering (RE). We are particularly interested by SoC in models that support the RE of software product lines (SPL).

    During RE, abstract models of the software to be developed and its environment are needed to tackle the complexity of the problem description. In SPL, this problem is even more serious since several (possibly hundreds) of software variants are developed at a time, bringing an additional source of complexity and the need for variability management.

    The supporting modelling languages that we study in relation to the RE of SPL are feature diagrams (FD) and problem frames (PF). We are interested in exploring SoC in, and through, FD and PF, and in their joint usage with other modelling languages.

Pascal Costanza

  • In Context-oriented Programming (COP), programs can be partitioned into behavioral variations expressed as sets of partial program definitions. Such layers can be activated and deactivated at runtime with dynamic scope in any order depending on the execution context. In previous work, we identified the need for application-specific dependencies between layers where the (de)activation of a layer requires the (de)activation of other layers, and suggested an efficient reflective interface for controlling such dependencies. However, that solution requires knowledge about complex low-level details of a particular COP implementation, which can be hard to master. We are currently looking into high-level approaches for enforcing such dependencies in a declarative ways, and have achieved some good first results using feature diagrams and the feature description language. We would like to use this work package to look further into this issue.

Ricardo Scandariato

  • As far as the decomposition track is concerned, DistriNet has defined a methodology leveraging security patterns as building blocks to define an architectural decomposition that fits the security requirements at hand.

    The methodology funds on the tenet that several competing software qualities (including security) must be considered in creating the architecture. Therefore, the methodology must support trade-off decisions by making them explicit and traceable. An initial approach to this problem is suggested, although it has to be improved on.

    Further, the methodology is supposed to be independent from the specific RE methodology that is selected to document the security requirements. However, we have tested it with scenario-driven approaches only. Further investigation is needed in order to assess what RE method fits better.
wp3/involved_partners.txt · Last modified: 2011/03/11 12:24 by eddy.truyen