Finding Incremental Solutions for Evolving
Requirements
Neil A. Ernst1, Alex Borgida2, John Mylopoulos1 and Ivan J. Jureta3
1 Department of Computer Science, University of Toronto
2 Department of Computer Science, Rutgers University
3 FNRS & Information Management, University of Namur
Abstract. This paper investigates aspects of the problem of software
evolution resulting from requirements change. In particular, while most
research on design for software focuses on nding some correct solution,
this ignores that such a solution is often only correct in a particular, and
often short-lived, context. Using a logic-based goal-oriented requirements
modeling language, the paper poses the problem of nding desirable so-
lutions as the requirements change. Among other possible criteria of
desirability, we consider minimizing the eort required to implement the
new solution, which involves reusing parts of the old solution. In general,
the solution of requirements problems is viewed as an exploration using
a \requirements engineering knowledge base" (rekb), whose specica-
tion is formalized. The paper reports on experience implementing the
rekb on top of a so-called \reason-maintenance system", and provides
evidence that incremental solution nding is indeed more ecient.
1 Introduction
It has long been recognized that much software development does not have a
denitive end. Systems which were assumed to last for only a few years are now
decades old, so-called \Browneld systems". And it has long been known that
the maintenance phase of a software system's lifecyle consumes the lion's share of
the resources, while consisting mostly of adaptive rather than corrective changes.
There are at least three major reasons for why the end result of software de-
velopment is indeterminate. First, software exists in a technological ecosystem
that has been rapidly changing throughout the history of computing (witness
the unexpected success of tablet computing paradigms). Second, the tools and
techniques we use to design and build software solutions are more powerful than
ever. Among others, agile project management has resulted in a new empha-
sis on incremental and iterative delivery. According to such methods, software
development is never `complete'. Finally, for many projects a common way to
elicit requirements is by reusing requirements from previous projects, which is
associated with lower requirements volatility [1].
The objective of this paper is to re-consider the problem of nding a solution
to the requirements problem in a goal-oriented requirements modeling and anal-
ysis framework, when the requirements are undergoing change. In particular, we