Modularizing early architectural assumptions in scenario-based requirements

9Citations
Citations of this article
16Readers
Mendeley users who have this article in their library.

This article is free to access.

Abstract

Early architectural assumptions (EAAs) are initial assumptions about the architectural solution that are made already during requirements elicitation. Such EAAs are inherently present when applying requirements engineering methods and techniques situated at the transition to architecture, for example those adhering to the Twin Peaks model to software engineering. In the current state-of-the-art, early architectural assumptions (EAAs) are documented implicitly, and they are tangled within and scattered across heterogeneous requirement artifacts. This makes it hard to distinguish EAAs from actual requirements, analyze their relevance, and bring them in relation to architectural decisions taken in later development stages. As a consequence, early development activities in the transition to architecture are hindered by the lack of explicit support for EAAs. In this paper, we address this problem in the context of scenario-based requirements (use cases and quality attribute scenarios). We present a system meta-model for EAAs, and provide an aspect-oriented requirements language that allows the instantiation of EAAs in terms of use case-level pointcuts. We employ our prototype implementation of above-mentioned techniques to evaluate and illustrate the benefits of making EAAs explicit in the early stages of development, specifically in terms of modularity and requirements navigability. © 2014 Springer-Verlag.

Cite

CITATION STYLE

APA

Van Landuyt, D., & Joosen, W. (2014). Modularizing early architectural assumptions in scenario-based requirements. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8411 LNCS, pp. 170–184). Springer Verlag. https://doi.org/10.1007/978-3-642-54804-8_12

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free