Current and Future Bots in Software Development
Paper in proceeding, 2019

Bots that support software development ("DevBots") are seen as a promising approach to deal with the ever-increasing complexity of modern software engineering and development. Existing DevBots are already able to relieve developers from routine tasks such as building project images or keeping dependencies up-to-date. However, advances in machine learning and artificial intelligence hold the promise of future, significantly more advanced, DevBots. In this paper, we introduce the terminology of contemporary and ideal DevBots. Contemporary DevBots represent the current state of practice, which we characterise using a facet-based taxonomy. We exemplify this taxonomy using 11 existing, industrial-strength bots. We further provide a vision and definition of future (ideal) DevBots, which are not only autonomous, but also adaptive, as well as technically and socially competent. These properties may allow ideal DevBots to act more akin to artificial team mates than simple development tools.

Sotware Engineering

Taxonomy

Software Bot

Author

Linda Erlenhov

Chalmers, Computer Science and Engineering (Chalmers), Software Engineering (Chalmers)

Francisco Gomes

University of Gothenburg

Riccardo Scandariato

University of Gothenburg

Philipp Leitner

Chalmers, Computer Science and Engineering (Chalmers), Software Engineering (Chalmers)

2019 IEEE/ACM 1st International Workshop on Bots in Software Engineering (BotSE)

7-11 8823643
978-172812262-5 (ISBN)

1st IEEE/ACM International Workshop on Bots in Software Engineering, BotSE 2019
Montreal, Canada,

Areas of Advance

Information and Communication Technology

Subject Categories

Software Engineering

DOI

10.1109/BotSE.2019.00009

More information

Latest update

7/18/2023