Views on models: a survey of existing approaches

When dealing with complex systems, information is very often fragmented across many different models expressed within a variety of (modeling) languages. To provide the relevant information in an appropriate way to different kinds of stakeholders, (parts of) such models have to be combined and potentially revamped by focusing on concerns of particular interest for them. Thus, mechanisms to define and compute views over models are highly needed.

We have proposed our own solutions to this problem, mainly EMF Views (a view mechanism to integrate heterogeneous models) but also, the simpler (and therefore, less expressive) EMF Profiles and EMF Facet approaches.

But many others exist. In fact, we have identified 16 tools to create model views.  For the creation of this catalog of view modeling approaches we have conducted a literature review on this topic. This survey has been performed by Hugo Bruneliere , Erik Burger, Manuel Wimmer and myself  and published in the SoSyM journal:  “A feature-based survey of model view approaches” (read it for free here).

In the paper, we provide a detailed overview of these 16 tools and the current state of the art in this area. As a result, we discuss the main capabilities of existing approaches and propose a corresponding research agenda. We notably contribute a feature model describing what we believe to be the most important characteristics of the support for views on models, some clarifications on the terminology in the domain (viewtype, viewpoint, view,…) and identify the following open research challenges in the domain: 1 – Terminology inconsistencies,  2- View updating problem, 3 – Incremental view maintenance, 4 – Concrete syntax generation for the view data and 5 – Security aspects (related to this point on security compliant model views).

We expect this work to be helpful to both current and potential future users and developers of model view techniques, as well as to any person generally interested in model-based software and systems engineering.

Featured image cropped from Photo by Erik Eastman on Unsplash