A Second Comprehensive Application Example - Model-Driven and Software Product Line Engineering - page 195

which are selected from the binding model and the pre-created
decision models.
7.4. Variability expression and product configuration
As presented previously, playing the role of product line
architects, we use metamodels and feature models as our base
core assets to capture and express variability; playing the role
of product designers, we configure products by creating models
that conform to metamodels and binding models.
7.4.1. Metamodels
In FieSta, we use a multi-staged approach for configuration
and derivation of products. To create the product line of
collection manager systems,we create metamodels playing the
role of product line architects. The first metamodel we built as
product line architects is the problem space metamodel , which
serves to define the domain concept we want to manage and
its characteristics. Problem Space models, which are created
by product designers, do not include GUI concepts or details of
functionality.OthermetamodelscontainGUIandfunctionality
concepts, which are orthogonal to the domain concept chosen
to be managed. Every metamodel plays a different role during
the product line development lifecycle. The first metamodel is
used by designers during the configuration process to create
domainmodels,whicharethestartingpointtoderivecollection
manager systems. The other two metamodels capture the
sources of variability that characterize our product line:
- Problem Space Metamodel. This metamodel includes
concepts regarding the domain concept to be managed and its
characteristics.
- Kernel Metamodel. Every system has functionalities
according to product user choices;the kernel metamodel allows
us to model variable products from the end-user functionality
point of view.