1 |
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - |
2 |
David Garcia, Marcelo Serrano Zanetti and Frank Schweitzer |
3 |
Chair of Systems Design – www.sg.ethz.ch – ETH Zurich |
4 |
|
5 |
|
6 |
= Abstract |
7 |
|
8 |
We analyze the relation between the emotions and the activity of |
9 |
contributors in the Open Source Software project Gentoo. Our case study |
10 |
builds on extensive data sets from the project’s bug tracking platform |
11 |
bugzilla, to quantify the activity of contributors, and its mail |
12 |
archives, to quantify the emotions of contributors by means of |
13 |
sentiment analysis. The Gentoo project is known for a considerable drop |
14 |
in development performance after the sudden retirement of a central |
15 |
contributor. We analyze how this event correlates with the negative |
16 |
emotions, both in bilateral email discussions with the central |
17 |
contributor, and at the level of the whole community of contributors. |
18 |
We then extend our study to consider the activity patters on Gentoo |
19 |
contributors in general. We find that contributors are more likely to |
20 |
become inactive when they express strong positive or negative emotions |
21 |
in the bug tracker, or when they deviate from the expected value of |
22 |
emotions in the mailing list. We use these insights to develop a |
23 |
Bayesian classifier that detects the risk of contributors leaving the |
24 |
project. Our analysis opens new perspectives for measuring online |
25 |
contributor motivation by means of sentiment analysis and for real-time |
26 |
predictions of contributor turnover in Open Source Software projects. |
27 |
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - |
28 |
|
29 |
http://arxiv.org/abs/1306.3612 |
30 |
|
31 |
|
32 |
Merry reading, |
33 |
jer |