Home > Research > Publications & Outputs > Angry-Builds

Links

Text available via DOI:

View graph of relations

Angry-Builds: An empirical study of affect metrics and builds success on GitHub ecosystem

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Published
  • Marco Ortu
  • Andrea Pinna
  • Roberto Tonelli
  • Michele Marchesi
  • David Bowes
  • Giuseppe Destefanis
Close
Publication date21/05/2018
Host publicationXP '18: Proceedings of the 19th International Conference on Agile Software Development: Companion
Place of PublicationNew York
PublisherAssociation for Computing Machinery (ACM)
Number of pages2
ISBN (print)9781450364225
<mark>Original language</mark>English
Event19th International Conference on Agile Software Development, XP 2018 - Porto, Portugal
Duration: 21/05/201825/05/2018

Conference

Conference19th International Conference on Agile Software Development, XP 2018
Country/TerritoryPortugal
CityPorto
Period21/05/1825/05/18

Conference

Conference19th International Conference on Agile Software Development, XP 2018
Country/TerritoryPortugal
CityPorto
Period21/05/1825/05/18

Abstract

Automatic and repeatable builds are an established software engineering practices for achieving continuous integration and continuous delivery processes. The building phase of modern software systems is an important part of the development process such that dedicated roles as "Release Engineer" are more and more required. Software development is a collaborative activity, and when multiple developers work on the same project, they will be changing a shared master development branch at overlapping intervals. This overlap occurs because developers create parallel branches for working and then merge these branches when features are completed. Continuous integration, CI, is a workflow strategy which helps ensure everyoneâĂŹs changes will integrate with the current version of the project. This activity allows developers to catch bugs and reduce merge conflicts. Improving the building process leads to higher productivity and therefore shorter time to market, but understanding or measuring such a delicate phase is a big challenge. Open Source Communities provide valuable empirical data such as GitHub an Travis CI. These repositories represent a golden mine containing important data which can help researchers understanding the process behind the manufacturing of a software artifact. By analyzing Travis CI logs, we can directly connect a particular build with the development process behind it, not only regarding code changes but also regarding human activities, such as discussions about the implementation of a specific feature or bug resolution. Thanks to this information we can analyze the social activities of the build process enabling us to apply the same approach used for the development process.