Advice to Service-Oriented Computing Research
            Students (Invited Talk)

                                 Marlon Dumas

  University of Tartu, Estonia and Queensland University of Technology, Australia
                              marlon . dumas @ ut.ee




Extended Abstract


Service-Oriented Computing (SOC) is a study field at the crossroads of several
other fields, such as software engineering, information systems, databases, dis-
tributed computing and Internet computing. It is also a rather dynamic field
since it is constantly being reshaped by ongoing industrial developments. As a
result, SOC is by no means an easy field to undertake a PhD project. This talk
will discuss a number of factors that PhD students in the field need to be mindful
of when scoping, planning and executing their project.
    One such factor is the importance of balancing the scientific requirement of
rigor with the aim of being relevant and impacting the practice of SOC. This
latter goal can be pursued by undertaking an applicability check [1] early on
during the PhD project in order to assess the importance and timeliness of the
problem and expected outcomes. This effort should be complemented with an
ongoing reflection on the accessibility and suitability of the proposed solution,
without neglecting the need to follow research methods that withstand scientific
scrutiny.
     Another factor to be considered is the importance of being validation-aware
throughout the PhD project, rather than postponing all the validation effort un-
til the end. This goal can be achieved in a two-pronged manner: (i) by preparing
the validatory work and collecting data right from the start of a PhD project;
and (ii) by alternating design and development steps with validation steps. Dif-
ferent validation methods are commonly used in the field, including prototype
implementation, case study, performance experiment, mathematical proof, user
experiment, survey. Each of them has its own trade-offs and scope of applica-
bility [2]. Factors worth consideration when selecting a validation method is the
type of research question, the type of research outcome, and more broadly, where
the PhD project stands in the spectrum between design and empirical science.
    Finally, when selecting and scoping a PhD project, one has to be mindful of
current agendas and trends in the field. These agendas and trends will be briefly
discussed during the talk.
References
1. M. Rosemann and I. Vessey. Toward Improving the Relevance of Information Sys-
   tems Research to Practitioners: The Role of Applicability Checks. MIS Quaterly
   32(1):1–22, March 2008.
2. M Shaw. What Makes Good Software Engineering Research. International Journal
   of Software Tools for Technology Transfer, 4(1):1–7, 2002.