Resolving governance disputes in communities: A study of software license decisions

• Author: Vivianna Fang He,Georg Krogh,Yash Raj Shrestha,Phanish Puranam
• Date: 01 October 2020
• Published date: 01 October 2020
• DOI: http://doi.org/10.1002/smj.3181

RESEARCH ARTICLE

Resolving governance disputes in communities:

A study of software license decisions

Vivianna Fang He

1

| Phanish Puranam

2

|

Yash Raj Shrestha

1

| Georg von Krogh

1

1

ETH Zurich, Zurich, Switzerland

2

INSEAD, Singapore, Singapore

Correspondence

Vivianna Fang He, ETH Zurich,

Weinbergstrasse 56-58, Zurich CH 8092,

Switzerland.

Email: fhe@ethz.ch

Funding information

Schweizerischer Nationalfonds zur

Förderung der Wissenschaftlichen

Forschung, Grant/Award Number:

169441

Abstract

Research summary: Resolving governance disputes is

of vital importance for communities. Gathering data

from GitHub communities, we employ hybrid induc-

tive methods to study discussions around initiation and

change of software licenses—a fundamental and poten-

tially contentious governance issue. First, we apply

machine learning algorithms to identify robust patterns

in data: resolution is more likely in larger discussion

groups and in projects without a license compared to

those with a license. Second, we analyze textual data to

explain the causal mechanisms underpinning these pat-

terns. The resulting theory highlights the group process

(reflective agency switches disputes from bargaining to

problem solving) and group property (preference align-

ment over attributes) that are both necessary for the

resolution of governance disputes, contributing to the

literature on community governance.

Managerial summary: Online communities play an

increasingly important role in how companies innovate

across organizational boundaries and attract talent

across geographic locations. However, online commu-

nities are no Utopia; disputes abound even (more)

when we collaborate virtually. In particular, gover-

nance disputes can threaten the functioning and exis-

tence of online communities. Our study suggests that

Received: 31 January 2018 Revised: 2 April 2020 Accepted: 2 April 2020 Published on: 24 June 2020

DOI: 10.1002/smj.3181

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and

reproduction in any medium, provided the original work is properly cited.

© 2020 The Authors. Strategic Management Journal published by John Wiley & Sons, Ltd. on behalf of Strategic Management Society.

Strat. Mgmt. J. 2020;41:1837–1868. wileyonlinelibrary.com/journal/smj 1837

governance disputes in online communities either

unfold as bargaining over which solution is better or

searching for a satisfactory solution. The latter is more

likely to reach a resolution, when there is common gro-

und. Companies interested in leveraging the power of

online communities should (a) identify or train certain

participants to transform endless bargaining into col-

lective problem solving and (b) foster shared knowl-

edge and value basis among participants through

recruitment and strong organizational culture.

KEYWORDS

community governance, dispute resolution, inductive theorizing,

machine learning, problem solving

1|INTRODUCTION

Communities are an increasingly important form of organizing (Adler, 2001; Adler &

Heckscher, 2017; Lee & Edmondson, 2017). Prominent examples come from domains as diverse

as open source software (OSS) (Dahlander & Frederiksen, 2012; Foss, Frederiksen, &

Rullani, 2016; O'Mahony & Ferraro, 2007; von Hippel & von Krogh, 2003), common pool

resources (CPR

1

) (Dietz, Ostrom, & Stern, 2003; Ostrom, 1990), scientific consortia (e.g., Human

Genome Project, Collins, Morgan, & Patrinos, 2003), and communities of practice within and

across firms (e.g., 3D printing design, Stanko, 2016).

Like any organization that draws on diverse resources and efforts to pursue a set of goals,

communities encounter disputes among their constituent individuals and coalitions (Deutsch &

Krauss, 1962; Jehn & Bendersky, 2003; March, 1962; see also Amit & Schoemaker, 1993;

Danneels, 2008). Dispute resolution is therefore a hallmark of effective and efficient community

governance (Ostrom, 1990). An organization's core governance principles—about the direction,

control, and coordination of its members—are meant (at least in part) to prevent the emergence

of destructive disputes, and when disputes do occur, to aid in their resolution (Gulati, Law-

rence, & Puranam, 2005; Gulati & Singh, 1998; March & Simon, 1958; Williamson, 1975).

However, disputes can also erupt around the very governance principles themselves. If not

resolved, disputes over fundamental principles such as the distribution of resources, rights, and

responsibilities, can derail the functioning or even jeopardize the existence of a community.

Disputes pertaining to core governance principles that affect the entire community (governance

disputes hereafter) differ substantially from disputes concerning operational matters (e.g., the

division and allocation of tasks, or management of resources) that typically affect only a subset

of community members at a time (henceforth operational disputes). Whereas theory and

research on mechanisms for resolving operational disputes are rapidly growing, studies on

1

A type of good consisting of natural or human-made resource systems (e.g., irrigation systems, forests, or fisheries) that

is characterized by nonexcludability (difficult to exclude actors from gaining utility from the resource) and

subtractability (use by one actor precludes use by another).

1838 HE ET AL.

governance disputes remain rare (Saeed, McDermott, & Boyd, 2017). The nascent state of

knowledge on how communities resolve governance disputes (Dietz et al., 2003) warrants

empirically driven, inductive investigation.

One plausible reason why research is still scant on governance disputes in communities is

that records of such disputes are often inaccessible. Those momentous occasions, when most if

not all members were involved in resolving disputes over governance principles remain largely

buried in the communities' history, not documented or available for research. It is also possible

that unresolved governance disputes removed those protocommunities from possible scholarly

scrutiny today. As a result, analysis of governance disputes may be difficult for researchers

working with data limited to functioning communities (Ostrom, 1990), since by virtue of their

existence, these communities have escaped or resolved significant governance disputes and

their counterparts that did not may no longer exist.

We leverage an unusual research opportunity found in the public discussions on GitHub, an

online platform hosting numerous open source software development (OSS) communities, to

explore the question of how communities resolve governance disputes. We aim to understand the

conditions that increase or decrease the likelihood of resolution, and the processes that underpin

differences in outcomes. Two factors motivate our choice of empirical context: (a) OSS communi-

ties are an important contemporary type of community for organizing economic activity

(Dahlander & Frederiksen, 2012; O'Mahony & Ferraro, 2007), and (b) they provide a unique win-

dow into the phenomenon of governance dispute resolution with a nondecaying electronic record.

Our exploration focuses on a particular type of governance dispute: disputes vis-à-vis the

software license that governs a given community's core software development project (Mar-

kus, 2007; O'Mahony, 2003). License disputes are of strategic importance for an OSS commu-

nity, because they determine how the software is used, modified, and combined with other

software(s) (Lerner & Tirole, 2005), thereby affecting all downstream users' innovation capabili-

ties (Murray & O'Mahony, 2007). Furthermore, they have significant impact on the

community's ability to attract not only developers (Lerner & Tirole, 2005; Subramaniam, Sen, &

Nelson, 2009) but also corporate partners, who then combine OSS products or services with pro-

prietary ones (Deodhar, Saxena, Gupta, & Ruohonen, 2012; Economides & Katsamakas, 2006).

We sample disputes surrounding the initiation or revision of a software license decision, col-

lecting quantitative and qualitative data on these disputes. The dispute instances we discovered

involve a large volume of data, making it difficult to rely solely on traditional inductive methods

that involve manually analyzing each case separately. Therefore, we combine machine learning

analysis with in-depth content analysis to develop a novel theory (Puranam, Shrestha, He, & von

Krogh, 2020; Tidhar & Eisenhardt, 2020). In effect, we pioneer a two-pronged approachto induc-

tive research: First, using machine-learning algorithms we document robust patterns across the

entire sample (183 cases). These patterns are unlikely to be the result of sampling error or func-

tional form idiosyncrasies, because the algorithms we apply vary in the functional forms and we

repeatedly split the data to build and test models using out-of-sample prediction accuracy as a

criterion. Second, we focus on analyzing in depth a randomly selected sub-sample (61 cases) to

enhance the theoretical understanding of mechanisms underlying the patterns we detected.

Our key finding is that governance disputes in OSS communities can take the form of either

bargaining or problem solving processes. The latter group process is more likely when “reflec-

tive agency”is manifested. This involves individual interventions that can help steer the discus-

sion toward the attributes of the software licenses (e.g., openness and permissiveness) rather

than the alternatives themselves (e.g., GPL and MIT). To the extent that the group has the prop-

erty that there is pre-alignment of preferences on attributes among members, attribute-based

HE ET AL.1839