Conclusion - Model-Driven and Software Product Line Engineering - page 240

enrich or extend existing programming languages to cope with
variability and to close the gap from variability to source code.
The next section summarizes the main concepts and
notations introduced in this topic. Then, the last section
outlines some issues of MD-SPL engineering discussed in the
previous chapters and concludes with some perspectives.
9.1. Topic summary
Onecanobservethatmanysoftwarecompaniesarebuilding
applications that share more commonalities than differences.
Theyoftenrepeatedlyaddnewfeaturesandbuildnewvariants
or releases of their applications. The strategy of systematic
and planned re-use of software artifacts was seen as a
means to improve software cost, productivity, and quality.
Software product line engineering is a new trend in software
development, which promotes the re-usability of artifacts.
Benefits of a product line are clear, provided that the initial
investment has been precisely measured. Engineers do not
have to minimize the complexity and the organization needed
tosuccessfullylaunchandmanageasoftwareproductline.This
engineering involves a complex two-fold software development
process: domain engineering and application engineering.
Domain engineering is responsible for defining the domain
scope and modeling the variability of the product line. This
is an important and critical engineering step where the core
assets are built and the production plan to derive products
is made. The variability model is a structured representation
of the commonalities and the variabilities of the products in
the scope of the line. Abstractly, a variability model is a base
descriptionwithvariationpointsandtheirassociatedvariants.
One of the main notations is feature modeling; however,
metamodels are increasingly used to represent variability.
Application engineering is the process of creating a product
fromtheproductconfigurationandthecoreassetsfollowingthe
production plan. The product configuration is an assembly of