Prof. Dr. Cesare Pautasso

An Empirical Basis for Software Architecture Research

Rick Kazman, Roberto Tonelli, Cesare Pautasso

Patrizio Pelliccione, Rick Kazman, Ingo Weber, Anna Liu (eds.)

Software Architecture - Research Roadmaps from the Community, Springer, pp. 87 - 100

2023

Abstract

Despite the clear need for and importance of performing empirical studies as part of software architecture research, there is still a lack of curated, standardized, clean, well-maintained, documented, easily accessible, reusable, and shared datasets. In this chapter, we provide an overview of the problems, of the motivations, and of the opportunities currently related to mining and sharing datasets for researchers in software architecture. We first explore and describe which artifacts should be included into such datasets, such as code, documentation, and requirements, but also including other architecturally relevant artifacts, such as architectural decision records, models, and other kinds of documentation. This information can be complemented with revision history logs, social metadata, and email or chat discussion archives. The availability of such datasets would enable not only architectural reconstruction studies but would also help to catalyze broader and more ambitious program of empirical studies in software architecture research.

Download

URL: https://link.springer.com/chapter/10.1007/978-3-031-36847-9_5

ISBN: 978-3-031-36846-2

DOI: 10.1007/978-3-031-36847-9_5

Citation

Bibtex

@inbook{sa:2023:empirical,
	author = {Rick Kazman and Roberto Tonelli and Cesare Pautasso},
	title = {An Empirical Basis for Software Architecture Research},
	editor = {Patrizio Pelliccione and Rick Kazman and Ingo Weber and Anna Liu},
	booktitle = {Software Architecture - Research Roadmaps from the Community},
	year = {2023},
	pages = {87 - 100},
	publisher = {Springer},
	organization = {Springer},
	abstract = {Despite the clear need for and importance of performing empirical studies as part of software architecture research, there is still a lack of curated, standardized, clean, well-maintained, documented, easily accessible, reusable, and shared datasets. In this chapter, we provide an overview of the problems, of the motivations, and of the opportunities currently related to mining and sharing datasets for researchers in software architecture. We first explore and describe which artifacts should be included into such datasets, such as code, documentation, and requirements, but also including other architecturally relevant artifacts, such as architectural decision records, models, and other kinds of documentation. This information can be complemented with revision history logs, social metadata, and email or chat discussion archives. The availability of such datasets would enable not only architectural reconstruction studies but would also help to catalyze broader and more ambitious program of empirical studies in software architecture research.},
	keywords = {software architecture},
	isbn = {978-3-031-36846-2},
	doi = {10.1007/978-3-031-36847-9_5},
	url = {https://link.springer.com/chapter/10.1007/978-3-031-36847-9_5}
}