reference any other. Because of that, it can be reused
in different contexts, such as in the modelling of
software or business processes.
If we think about the modelling reuse level, the
processes, metrics and even questionnaires modelled
for a given experiment using our DSLs can be
completely or partially reused in the context of other
experiments. Hence, despite reuse has been not
explicitly investigated in this study, there is a great
opportunity to explore the reuse of the specification
of an experiment using our DSLs.
Variability in DSLs. In the Configuration
Knowledge Experiment modelling, there were three
processes related to the experiment, where each one
is related to a specific SPL implementation, which
represents one factor of the experiment. During the
workflows generation, these processes will vary
conform the randomized selected SPL for the
treatment, thus it represents a variation point. The
identification of these variabilities is important in
our approach in order to support their specification
and customization. In the case of our Experiment
Software Engineering DSLs, for example, we plan to
explicitly specify such variabilities in the DSLs in
order to support the customized generation of
workflows for each subject according to the
experiment statistical design, for example, Latin-
square.
4 CONCLUSIONS
This paper investigated the composition DSLs
problems through an exploratory study using a
proposed method. Our study focuses on the
application of the method for the composition of
Ecore-based DSLs implemented using the xText
framework. The composition method was applied in
the modelling and composition of DSLs that allow
specifying and executing controlled experiments in
the experimental software engineering domain. Our
main contributions were: (i) the evaluation of the
investigated method in a new context comparing to
the previous one; and (ii) the two experiments
specification using the DSLs composition that
supports the modelling of different perspectives of a
controlled experiment.
As future work, we intend to extend our model-
driven approach to completely support the workflow
generation. Furthermore, we will investigate
techniques and mechanisms to explicitly model
variabilities in these DSLs in order to address the
customized generation of workflows for subjects
according to the chosen experimental statistical
design.
ACKNOWLEDGEMENTS
This work was partially supported by the National
Institute of Science and Technology for Software
Engineering (INES, www.ines.org.br), funded by
CNPq under grants 573964/2008-4, 560256/2010-8,
and 552645/2011-7.
REFERENCES
Bézivin, J. and Jouault, F., 2005. Using ATL for Checking
Models. Workshop GraMoT. Tallinn: pp. 69-81.
DSL Composition, 2012. [Online] Available at: https://
sites.google.com/site/compositiondsl/
Cirilo, E. et al, 2011. Configuration Knowledge of
Software Product Lines: A Comprehensibility Study.
Workshop on VariComp. New York: pp. 1-5.
Clements, P. and Northrop, L., 2011. Software Product
Lines: Practices and Patterns. Addison-Wesley.
Czarnecki, K. and Helsen, S., 2006. Feature-based survey
of model transformation approaches. IBM Systems
Journal - MDSD, 45(3), pp. 621-645.
Deelstra, S. et al, 2005. Product derivation in software
product families:a case study. JSS, 74(2), pp. 173-194.
Freire, M. A. et al, 2011. Automatic Deployment and
Monitoring of Software Processes: A Model-Driven
Approach. SEKE 2011, 9 dec., pp. 42-47.
Freire, M. A. et al, 2012. Software Process Monitoring
Using Statistical Process Control Integrated in
Workflow Systems. SEKE 2012, 20 jan., pp. 557-562.
Hessellund, A. and Lochmann, H., 2009. An Integrated
View on Modeling with Multiple Domain-Specific
Languages. IASTED on ICSE. pp. 1-10.
Hessellund, A., 2007. SmartEMF: guidance in modeling
tools. OOPSLA. New York: ACM, pp. 945-946.
Hessellund, A., 2009. Domain-specific multimodeling,
Denmark. Thesi.
Hessellund, A. et al, 2007. Guided Development with
Multiple Domain-Specific Languages. MoDELS’2007,
Nashville, Springer, pp. 46-60.
Lochmann, H. and Bräuer, M., 2007. Towards Semantic
Integration of Multiple Domain-Specific Languages
Using Ontological Foundations. MoDELS, Nashville.
Lochmann, H.; Grammel, B., 2008. The Sales Scenario: A
Model-Driven Software Product Line. In:: Software
Engineering (Workshops). s.l.:s.n., pp. 273-284.
Mens, T. et al, 2006. Detecting and resolving model
inconsistencies using transformation dependency
analysis. MoDELs. Genova: Springer, pp. 200-214.
Nentwich, C., et al, 2003. Consistency management with
repair actions. ICSE. Portland: IEEE, pp. 455-464.
Wohlin, C. et al, 2000. Experimentation in Software
Engineering: An Intoduction. Norwell: Kluwer
Academic Publishers
MODELSWARD2013-InternationalConferenceonModel-DrivenEngineeringandSoftwareDevelopment
156