TY - GEN
T1 - More common than you think
T2 - 23rd IEEE International Conference on Software Analysis, Evolution, and Reengineering, SANER 2016
AU - Pinto, Gustavo
AU - Steinmacher, Igor
AU - Gerosa, Marco Aurélio
N1 - Publisher Copyright:
© 2016 IEEE
PY - 2016/5/20
Y1 - 2016/5/20
N2 - Source code hosting websites (code forges) have recently changed to more social environments, and the contribution process evolved to the so-called pull-based development model. Due to the facilities brought by this evolution, Open Source Software (OSS) projects are now facing a high exposure, leading to an increasing number of contributors. However, not all these contributors want to have a long-term engagement with the project. In fact, popular projects are known to have a restricted set of core developers who drive the project, but now these projects count on a broad set of “not that involved” developers, which are responsible for a long tail of small contributions. In this paper, we shed the light on this important but overlooked set of developers: the casual contributors (also known as drive-by commits). First, we mined popular software repositories hosted on GitHub to investigate how common casual contributions are, and what are their characteristics. Second, we conducted two surveys with (1) the casual contributors and (2) the project maintainers aimed at understanding what motivates casual contributors and how they are perceived. Our results showed that although casual contributors are rather common (48.98% of the whole population of contributors in the projects analyzed), they are responsible for only 1.73% of the total number of commits. We also found that casual contributions are far from being trivial: even though a significant proportion of them are fixing typos and grammar issues (28.64%), we found several of them that have fixed bugs (30.20%), added new features (18.75%), and refactored code (8.85%). Still, we found that both casual contributors and project maintainers believe that casual contributions have more benefits than drawbacks. As a casual contributor said: “every bit helps”.
AB - Source code hosting websites (code forges) have recently changed to more social environments, and the contribution process evolved to the so-called pull-based development model. Due to the facilities brought by this evolution, Open Source Software (OSS) projects are now facing a high exposure, leading to an increasing number of contributors. However, not all these contributors want to have a long-term engagement with the project. In fact, popular projects are known to have a restricted set of core developers who drive the project, but now these projects count on a broad set of “not that involved” developers, which are responsible for a long tail of small contributions. In this paper, we shed the light on this important but overlooked set of developers: the casual contributors (also known as drive-by commits). First, we mined popular software repositories hosted on GitHub to investigate how common casual contributions are, and what are their characteristics. Second, we conducted two surveys with (1) the casual contributors and (2) the project maintainers aimed at understanding what motivates casual contributors and how they are perceived. Our results showed that although casual contributors are rather common (48.98% of the whole population of contributors in the projects analyzed), they are responsible for only 1.73% of the total number of commits. We also found that casual contributions are far from being trivial: even though a significant proportion of them are fixing typos and grammar issues (28.64%), we found several of them that have fixed bugs (30.20%), added new features (18.75%), and refactored code (8.85%). Still, we found that both casual contributors and project maintainers believe that casual contributions have more benefits than drawbacks. As a casual contributor said: “every bit helps”.
UR - http://www.scopus.com/inward/record.url?scp=85053010947&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85053010947&partnerID=8YFLogxK
U2 - 10.1109/SANER.2016.68
DO - 10.1109/SANER.2016.68
M3 - Conference contribution
AN - SCOPUS:85053010947
T3 - 2016 IEEE 23rd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2016
SP - 112
EP - 123
BT - 2016 IEEE 23rd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2016
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 14 March 2016 through 18 March 2016
ER -