A classification of value for software architecture decisions

5Citations
Citations of this article
24Readers
Mendeley users who have this article in their library.
Get full text

Abstract

This paper introduces a classification for decisions originating from work performed by architects. With the creation of a new architecture, all observed decisions were documented using an existing taxonomy extended with the introduced classification. In the first four months, 80 decisions were documented. Not all decisions have the same value for the architecture and one needed a classification to reason about importance of decisions. After realization of the first increment of the architecture a sanity check was performed: The architects showed how the six most important design artefacts and the fifteen most important architectural constraints and prerequisites were related. The relationship was via decisions and the classification helps to reduce the work to make and maintain this connection over time. The classification is dynamic and over time decisions can be classified differently. This enables architectural learning by pointing out which decisions were taken too early or had little impact. © 2010 Springer-Verlag Berlin Heidelberg.

Cite

CITATION STYLE

APA

Eklund, U., & Arts, T. (2010). A classification of value for software architecture decisions. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6285 LNCS, pp. 368–375). https://doi.org/10.1007/978-3-642-15114-9_30

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