| Phase | 6-Grad |
| Prog. | MSc PM |
| Adm. – Grad. | 2022 – 2024 |
| Dir.; Codir. | Stéphane Gagnon |
| <TBC> | |
| UQO | https://di.uqo.ca/id/eprint/1725/ |
Leaders of Open-Source Projects: Explaining Participation from Perceived Challenges and Opportunities
Zare-Charaki, Maryam
Free/Libre Open-Source Software (FLOSS) projects are mostly inspired by the Open Innovation and Agile Projects movements. They present significantly different leadership challenges than more traditional software development methodologies, and debates continue as to how to identify the best professional profile and focus to lead FLOSS projects.
We will review some considerations related to leadership initiative within FLOSS project. Specif-ically, we test the following hypotheses for research questions 1 to 5: H1: Professionals involved in both proprietary and open-source projects have a higher level of involvement in FLOSS leader-ship compared to those involved only in one type of project; (H2) Professionals who perform both development and contribution roles have a higher level of involvement in FLOSS leadership com-pared to those who perform only one of these roles; (H3) There is a significant difference in the level of involvement in FLOSS leadership between male and female professionals; (H4) Profes-sionals involved in both FLOSS and proprietary projects perceive higher career success, in terms of monthly income and job satisfaction, compared to those involved only in one type of project; (H5) Members involved solely in FLOSS projects have different engagement levels compared to those participating in both FLOSS and proprietary projects, with variations segmented by their roles as non-technical contributors, developers, or both.
Using the stepwise (backward elimination) regression algorithm in the statistical software Tana-gra, we develop models to explain the number of FLOSS projects under leadership, explained by a set of 22 items measuring 5 constructs of Perceived Challenges, Opportunities, and Career Ben-efits. We segment our dataset as well using 3 control variables: whether leaders are involved in FLOSS + Proprietary vs. Pure FLOSS; the role members have in FLOSS: non-technical contribu-tors, developers, or both; and gender. Preliminary results show that our 9 models have R^2 between 0.83 and 0.97, reducing from 22 to between 2 to 7 highly significant explanation (p values mostly around 0.00 and a few up to 0.05). These findings show that the different sub-groups of FLOSS project members have highly different behavior (i.e., explanation and coefficients) associated withtheir respective FLOSS leadership involvement, and indicate which ones are most associated with key constructs of Perceived Challenges and Opportunities.
This research also offers an in-depth examination of the dynamics between involvement in FLOSS and proprietary software projects considering these findings. These insights are particularly note-worthy given that at least half of the members in FLOSS projects are concurrently engaged in proprietary software projects. There is still debate as to the relative effectiveness of this profes-sional strategy, as well as its value for the FLOSS community. Using a Welch’s t-test in Excel, we perform a comparative study of these 2 groups (FLOSS + Proprietary vs. Pure FLOSS), further segmented by the role members have in FLOSS: non-technical contributors, developers, or both (2 X 3 = 6 sub-groups + 2 groups = 8 cells = 4 pairs). Preliminary results on two constructs of four items, career success and FLOSS engagement, show fifteen significant (0.00 to 0.05 and one 0.10) p values out of thirty-two comparisons (4 pairs X 4 items X 2 constructs). These results confirm that more participation in proprietary software is associated with higher monthly income, higher job satisfaction, and higher job autonomy, but with less FLOSS engagement.
Our analysis showed several critical factors influencing leadership involvement within the FLOSS community. Time allocation (FE1) and active contribution to multiple projects (FE3) appeared as significant explainers of leadership roles, highlighting the importance of engagement and contri-bution depth. It is remarkable that job satisfaction (CS2) is likewise essential, suggesting that pro-fessional contentment is intricately linked to leadership involvement. On the other hand, legal is-sues (PC4) and software functionality uniformity (SC2) were deterrents to leadership, underscor-ing the need for mechanisms to address these challenges.
Additionally, dual participation in both FLOSS and proprietary projects also improves the chance to work in managerial positions, as well as being correlated with a higher level of monthly income. The study also emphasized the importance of attracting users and contributors (TC1) for aspiring leaders, showing the complex interplay of factors driving leadership in FLOSS. This foreground-ing of these key drivers afforded our work with valuable perspectives of what those dynamics of leadership consisted of within the hybrid space of FLOSS and proprietary software development.
We found that not just amount of time contributed to FLOSS projects but also the number of pro-jects contributed to as well as effective communication within these projects are significant factors for achieving leadership positions, especially for women. Women who invest more time and con-tribute to a variety of projects tend to assume more leadership roles. However, fundraising diffi-culties are a significant barrier, one that may deter women from transitioning into leadership. Ef-fective project communication is essential, as women who excel in internal project communication are more likely to ascend to leadership roles, highlighting the importance of soft skills alongside technical contributions. This further emphasizes the need for targeted support on how these chal-lenges can be addressed enabling the environment to promote women leadership in the FLOSS community.