The (C)omprehensive (A)rchitecture (P)attern (I)ntegration method: Navigating the sea of technology

TL;DR

This paper introduces CAPI, a decision tree-based method for guiding architectural pattern selection in software projects, reducing complexity and aiding both academic and industry users in making informed technology choices.

Contribution

CAPI provides a novel, structured approach to architectural pattern selection using a diagnostic decision tree, improving decision-making in complex technological landscapes.

Findings

CAPI is perceived as helpful by users.

CAPI can replicate productive architectural environments.

Technology selection is often trial-and-error.

Abstract

The technological landscape changes daily, making it nearly impossible for a single person to be aware of all trends or available tools that may or may not be suitable for their software project. This makes tool selection and architectural design decisions a complex problem, especially for large-scale software systems. To tackle this issue, we introduce CAPI, the Comprehensive Architecture Pattern Integration method that uses a diagnostic decision tree to suggest architectural patterns depending on user needs. By suggesting patterns instead of tools, the overall complexity for further decisions is lower as there are fewer architectural patterns than tools due to the abstract nature of patterns. Moreover, since tools implement patterns, each non-proposed pattern reduces the number of tools to choose from, reducing complexity. We iteratively developed CAPI, evaluating its understandability and usability in small studies with academic participants. When satisfied with the outcome, we performed a user-study with industry representatives to investigate the state-of-the-art in technology selection and the effectiveness of our proposed method. We find that technology selection is largely performed via trial and error, that CAPI is uniformly perceived as helpful, and that CAPI is able to reproduce the productive architectural environments of our participants.