Requirements as Goals and Commitments Too - Intentional Perspectives on Information Systems Engineering - page 137

ones, and are regulated and constrained by internal organizational rules, business
processes, external laws and regulations [ 15, 31] . Among the challenging prob-
lems related to the analysis and design of a socio-technical system is the problem
of understanding the requirements of its software components, the ways technol-
ogy can support human and organizational activities, and the way in which the
structure of these activities is influenced by introducing technology. In particular,
in a socio-technical system, human, organizational and software actors rely heav-
ily on each other in order to fulfill their respective objectives. Not surprisingly, an
important element in the design of a socio-technical system is the identification of
a set of dependencies among actors which, if respected by all parties, will fulfill all
stakeholder goals, the requirements of the socio-technical system.
This chapter is structured as follows. Section 2 provides a comprehensive
overview on commitments, specifically on their usage in multiagent systems.
Section 3 illustrates how commitments can be used with goals to specify require-
ments, and introduces some reasoning principles. Section 4 exemplifies how the
reasoning may be applied in a travel agency setting. Section 5 compares our model
to related work. Finally, Sect. 6 concludes with a summary of our approach.
2 Commitments in Multiagent Systems
The concept of commitment spans many disciplines, from Philosophy of Mind, to
Psychology, Sociology and Economics. A review of the literature suggests that the
concept has only been studied in the later half of the last century (it is true: Aristotle
did not discover everything!).
Commitments as a computational abstraction have a long history in Computer
Science. Bratman [ 3] and Cohen and Levesque [ 9] formulated the notion of an
agent's commitment to his intentions. Singh [ 33] labeled commitments of this nature
as psychological commitments, and instead stressed the notion of social commit-
ment , that is, commitments among agents. In particular, Singh showed that social
commitments are key to modeling communication among agents [ 34] , and conse-
quently to the development of large systems consisting of autonomous, interacting
entities—in other words, multiagent systems. In the following, the term commitment
is used solely in the sense of a social commitment .
Singh [ 35] also elucidated the key ontological aspects of commitments. Since
then, commitments have been applied as a basis for flexible interaction among
agents [ 41, 42] ; towards the formulation of agent communication languages [ 17] ,
as an abstraction for business process design [ 11, 14] ; towards a type theory for
protocols [ 6, 24] ; towards understanding interoperability among agents [ 6, 7] ; and
towards formulating a service-oriented architecture [ 38] . Aspects related to reason-
ing about commitments have been addressed in [ 7, 13, 16, 36] . Commitments have
also been recently applied in requirements engineering [ 39] , and for monitoring in
conjunction with goals [26] .
Below, we characterize multiagent systems especially emphasizing the value of
commitments.