<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>Entering an ecosystem: The hybrid OSS landscape from a developer perspective</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Hanna Maenpaa</string-name>
          <email>hanna.maenpaa@helsinki.fi</email>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Fabian Fagerholm</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Myriam Munezero</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Terhi Kilamo</string-name>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Tommi Mikkonen</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>Department of Computer Science, University of Helsinki</institution>
          ,
          <addr-line>Helsinki</addr-line>
          ,
          <country country="FI">Finland</country>
        </aff>
        <aff id="aff1">
          <label>1</label>
          <institution>Tampere University of Technology</institution>
          ,
          <addr-line>Tampere</addr-line>
          ,
          <country country="FI">Finland</country>
        </aff>
      </contrib-group>
      <pub-date>
        <year>2017</year>
      </pub-date>
      <fpage>127</fpage>
      <lpage>137</lpage>
      <abstract>
        <p>Hybrid Open Source Software projects are virtual organizations that express characteristics of both static and dynamic behavior. They are choreographed through complex organizational structures that mix centralized governance with distributed community drivenness. While many communities use standard software tools to support their development processes, each community has its own ways of working and invisible power structures that in uence how contributions are submitted, how they are veri ed and how decisions about the long-term direction of the software product are made. Navigating this environment is especially challenging for new developers who need to prove their abilities to gain rights to make contributions. This paper provides a viewpoint on the factors that in uence a new developer's perception of the hybrid OSS developer community landscape. We apply an established developmental theory to build an initial model for the developer's context and discuss the model's validation, providing its practical and theoretical implications for building and managing on-line developer communities.</p>
      </abstract>
      <kwd-group>
        <kwd>Open Source</kwd>
        <kwd>community</kwd>
        <kwd>management</kwd>
        <kwd>on-boarding</kwd>
        <kwd>developer experience</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>1 Introduction</title>
      <p>Open Source Software (OSS) products are built collaboratively by developers
that jointly form a developer community. These communities are geographically
distributed, and they use virtual work environments, software tools and on-line
collaboration platforms, along with mutually agreed policies and practices for
developing their software product as a common good [Kil14]. Developer
communities can comprise of independent individuals or representatives of companies, or
in the case of "hybrid" communities: both [GBICMR13]. In hybrid communities,
development decisions are in uenced by versatile goals of individuals and often
competing commercial interests [TRGB15]. Therefore, ensuring that
stakeholders have equal opportunities to contribute is key for sustaining the reciprocity
of the collaboration and the motivation of the community's members.</p>
      <p>Even though individuals exist within the same context in developer
communities, they may perceive their environment in di erent ways. This is especially
true for new developers, who are yet to learn about the culture, customs and
realities of the community they want to join. So far, research has been performed
on how developer communities should be organized to accommodate
participation [DGO11, WO08], what factors help entry and on-boarding of its
members [FGMB14, FSBM14], and which traits healthy and welcoming communities
should display [Jan14, WJ07]. However, connections in between the community
context and the developer's experience have largely been unexplored, due to
e.g. the lack of a common vocabulary in describing the community phenomena
[AdOJ17]. In this paper, we address this gap by linking current literature on
developer community context and management to the viewpoint of an
individual developer by using an established theory of environmental factors in human
development [Bro77]. By studying the \Where, Who, and What" of the Hybrid
Open Source developer community context, we aim to lay ground for new
understanding on how hybrid developer communities can be built and managed to
be welcoming for new developers.</p>
      <p>This article is structured as follows. Section 2 rst reviews how the hybrid
OSS community context can be understood based on recent literature. Section 3
provides an overview of the Ecological Systems Model of Bronfenbrenner [Bro77],
and describes its application to the hybrid OSS developer community context.
Section 4 discusses and concludes the work.</p>
    </sec>
    <sec id="sec-2">
      <title>2 The community context</title>
      <p>One of the rst models to describe online developer communities was the onion
model of Nakakoji et al. [NYN+02]. The model focused on the developers with
the most in uence on the source code, establishing the terms "project leaders"
and "core developers", who base their actions on implicit knowledge about the
roles of the group and trust their peers in "just doing what needs to be done"
[MFH02]. In Nakakoji's model, role of "co-developers" was di erentiated to
active developers and peripheral developers who delivered source code updates,
yet were not in an active role in decision-making. Within this group, distributed
communication of the developers increased requirements for work coordination,
quality assurance and transparent, merit-based accreditation [MFH02]. The
exterior layer of the model became formed by passive software users, community
bystanders and bug reporters who had no direct contact with the source code
[NYN+02]. While inviting new contributors and sustaining them contributes to
the sustainability of a project [Jan14], it has long been preferable that new
members enter the community through the outer layers and proceed towards the core
based on their gained merit [YK03, NYN+02].</p>
      <p>Today, OSS developer communities can be enveloped in an ecosystem of third
party bene ciaries, who build their businesses on the OSS software produced by
the community. These stakeholders can interact through shared socio-technical
systems [FBACF17], taking approaches towards the developer community that
can vary from symbiotic to opportunistic behaviors [DM05]. The commercial
actors do not necessarily need to climb up the meritocracy ladder, yet they
can e.g. deploy developers to the project or be granted special privileges in its
decision-making based on partnership models [SWvK15]. As the stakeholders
can be erce competitors on a shared market [TRGB15], they can be forced
to form strategic alliances to advance their mutual goals such as stability or
interoperability of the software [FBACF17]. These invisible power structures are
hard to perceive, yet they can in uence priorities of development and decrease a
new developer's say in the project. Here, a well documented governance model,
along with transparent and consistent management practices can help to increase
equality among contributors [BJ12].</p>
      <p>As the hybrid environment with mixed goals and motivations grows in size, a
commercial or foundation-based central organization ("community sponsor") can
be established to ensure the community's sustainability by orchestrating its work.
To support its operation, the central orchestrator can deploy paid contributors
to the project [WO08] and sponsor and manage the community's development
tools, allowing it to control the development process to a degree. The orchestrator
can also acquire funding, recruit new developers and arrange marketing- and
community-building activities, which play an indistinguishable role in growing
the community and ensuring its health [MH13]. A community ecosystem needs to
endure changes and sustain its productivity in changing settings [MH13]: health
can be seen as a mix of the community's internal performance, its vigor and
its resilience to external disruptions [WJ07]. To support these aims, it is vital
to understand how new developers need to be supported to become productive
members and increase their participation in the community's activities.</p>
      <p>A key element in the hybrid OSS community phenomenon is its governance:
the way an organization is managed, including its powers, responsibilities and
decision making practises [DK09]. Governance principles include how rights to
participate are managed and how responsibilities are allocated amongst the
community's members [Jan14]. This includes the assignment of roles and
decisionmaking rights, as well as the measures and policies that enable continuous
assessment of the developer organization's performance and well-being. When the
community operates in a transparent way, its members can feel they can
understand the community's state and have an opportunity to in uence development
activities and processes. Restricting autonomy of the open developer
community may endanger the developers' motivation. This is especially true in hybrid
projects where the governance structure can a ect the motivation of outside
volunteers' participation [Sha06] and thus, also the uptake of the software.</p>
      <sec id="sec-2-1">
        <title>2.1 A focus on the developer</title>
        <p>A software developer's experience is tied closely to the project's development
tools and infrastructure, the developers' feelings about their work and the sense
of value that they see for their contributions [FSBM14, KM06]. In a healthy
community, members should be able to achieve su cient returns for their work,
to satisfy their needs and be encouraged to make new contributions [WJ07].</p>
        <p>Here, individual work tasks and their relevant knowledge become an essential
Discipline</p>
        <p>De nition of context
Source
Management Circumstances, conditions, situation or environ- Welter [Wel11].
science ments that are external to a speci c phenomenon</p>
        <p>and that enable or constrain it.</p>
        <p>Organization Situational opportunities and constraints that af- Johns [Joh06].
behavior fect the occurrence and meaning of organizational
behavior as well as functional relationships
between variables.</p>
        <p>Contextaware
puting</p>
        <p>Location of use, the collection of nearby people Schilit [SAW94].
com- and objects as well as the changes to those objects</p>
        <p>over time.</p>
        <p>Software Rich and complex network of elements across dif- Antunes et al.
development ferent dimensions such as personal, project, orga- [ACG10], De
nization, domain, product, teams, etc. Araujo et al.
[DSB+04].</p>
        <p>Linguistics</p>
        <p>Meaning surrounding intra-textual clues that may Chin [Chi94].</p>
        <p>be based on or be outside of the text itself.</p>
        <p>Hybrid OSS Combination of open community-driven devel- Sharma [SSR02].
development opment and company internal development
procommunities cesses.</p>
        <p>OSS commu- A thematic division of developers into core and Nakakoji et
nities periphery based on frequency of contribution and al. [NYN+02],
nature of their contributions. Mockus et al.
2002 [MFH02],
part of the context [KM06]. In the online developer community realm, this
context expands to a wide variety of freely available information and online tools
such as the versioning system, work ow coordination and quality management
tools, which together with discussion forums, chats and mailing lists make
distributed coordination and collaborative decision-making possible. Interactions
amongst the community's members take place in both formal meetings and
informal, spontaneous online commentaries. In the hybrid environment, a part of
the discussions and the project's decision-making take place outside the reach
of the open community. This can limit the sense of own possibility to in uence
developments and hamper meaningful social interactions, which are known to
be major barriers for the entry of newcomers to OSS projects [SSGR15].</p>
        <p>In OSS communities, newcomers are required to prove their ability and
motivation to make meaningful contributions to the project before they can
acquire rights to make changes to the software source code and participate in
decision-making about the software product's future [SSR02]. To achieve this,
new developers need to grasp not only the touch points for participation,
meaning of the software's licensing and governance processes, yet they also need to
understand the social dynamics and inherent conventions of the community.
Inherently, the organization of the development process, access to decision-making
and tasks, along with transparency of knowledge and processes form an
important leverage for managing contributions[WO08]. Successful entry to a
community is greatly dependent on the community's ability to provide this
information and mentor new contributors to achieve an active role, which in part
relies on a smooth learning process for new developers and their swift
integration to the socio-cultural values, work processes and prevailing practices of the
community[FSBM14]. Hence, designing and managing the entry process of new
developers requires understanding of the nature of the community context and
its re ection on the new developer's experience.</p>
      </sec>
    </sec>
    <sec id="sec-3">
      <title>3 Modeling the context</title>
      <p>The hybrid OSS ecosystem is a complex context, which we chose to explore
through the lens of the Bioecological model of Bronfenbrenner (1979) that
explains personal learning as "a function of forces that originate from the versatile
settings amongst which the individual is a part of" [Bro79]. The model organizes
environmental factors as layers similar to that of the Onion Model, positing the
individual human being with his or her personal characteristics in the epicenter.
Bronfenbrenner's view emphasizes a person's learning as an experiental process
to which a person's most proximate interactions have the most in uence. The
approach can be used to divide the hybrid OSS ecosystem context to four layers:
its micro-, meso-, exo- and macrosystem.</p>
      <p>The microsystem level illustrates phenomena of practical everyday life: a
person's most immediate surroundings and human contacts. More distant
phenomena such as infrequent contacts, local culture and its institutions are positioned
to the exosystem level. The layer in between, mesosystem, represents interactions
between the micro- and exosystems, emphasizing the synergistic nature of the
phenomena. As the exterior layer of the model, Bronfenbrenner describes the
concept of a macrosystem, which deals with the values, events and expectations
of the surrounding culture and society at large.</p>
      <p>Bronfenbrenner denotes that all the layers have symbiotic relationships with
each other and as for the ever-evolving nature of reality, are subject to changes in
time (chronosystem). We used this thinking to depict the hybrid OSS ecosystem
context in Figure 1. The next chapter, along with Table 2 elaborates the elements
in our proposal in detail.</p>
      <sec id="sec-3-1">
        <title>3.1 The example application</title>
        <p>We used examples from literature presented in Section 2 to deploy the di erent
layers of the Bioecological model. In our application (Figure 1), the individual
software developer is placed as the focal point: as both a perceiver of the hybrid
ecosystem context { and the subject of action and related personal development.</p>
        <p>Each individual has her personality, competencies and preferences that re ect to
the ways in which the person sees the community's opportunities for interaction.
The experience can be divided into cognitive, conative and a ective aspects
which guide her reactions to her surroundings [FGMB14]. Around the individual
exists a microsystem of direct in uences such as the developer's contributions to
the project, personal responsibilities, along with her interactions with peers and
shared knowledge repositories such as the source code, online documentation
and discussions.</p>
        <p>Interactions with the microsystem phenomena take place in face to face
settings, or as typical in online communities: through synchronous or asynchronous
personal interactions with the help of socio-technical tools and platforms. The
microsystem's factors enable, allow and invite engagement, as well as inhibit
participation and complicate proximate processes of the individual with her
immediate environment [BM06]. The exosystem layer gathers phenomena that are
Stakeholders Requirements engineers, code authors, committers, re- [NYN+02]
viewers and maintainers. Own team members, mentors [FSBM14]</p>
        <p>Individual</p>
        <p>Microsystem
Own competencies, goals, motivations and feelings.
[FGMB14]
Own knowledge, tasks, contributions, responsibilities, [KM06]
own status/rank as a community member. [SSR02]
[Jan14]
Software source code. Personal development tools and [Kil14]
common platforms: requirements management systems,
code review/testing and continuous integration tools,
shared knowledge repositories.</p>
        <p>Mesosystem: (the connecting layer)
Organization Code submission-, review-, testing-, integration- and [WO08]
of production release processes. Collaborative relationships amongst [NYN+02]
community members.</p>
        <p>Exosystem
Architecture of the software product. Rules and con- [SSR02]
ventions. Licensing and IPR strategy. Organization and [WO08]
governance of the community and its orchestrator. [DK09]
Stakeholders Di erent project teams and sub-communities. Project [NYN+02]
lead and high-level decision-making organizations, soft- [FBACF17]
ware users. Goals of partner organizations and keystone
players.
related to the community's internal culture and governance model, which
indirectly in uence the individual's environment and the way in which she engages
with her microsystem. These two are connected by the mesosystem: the
community's processes, including the di erent manual and automated stages of the
software development process, along with its decision-making- and accreditation
mechanisms. These de ne the community in terms of its structures, processes and
events that occur outside the developing individual's immediate surroundings.
The macrosystem comprises of the external environment that surrounds the OSS
project and exists independently from it. However, it in uences the exosystem
and subsequently: the individual's behavior. While the Free and Open Source
Software (FLOSS) ideology is emblematic to the environment, other
prevailing phenomena include the technological landscape, prevailing business models,
stakeholder alliances, industry- and market trends.</p>
      </sec>
    </sec>
    <sec id="sec-4">
      <title>4 Discussion and Summary</title>
      <p>The Open Source development context is traditionally viewed through the
notions of the Onion model and the core-periphery distribution of community
members. We extend this de nition by proposing a new application of an
established theory on external factors in human development by Bronfenbrenner
[Bro79] into the hybrid OSS ecosystem environment. We do this to build initial
understanding on how the software ecosystem context can be perceived by a
new developer that wishes to enter the community. Several factors may limit
validity of our proposal. Firstly, Bronfenbrenner's Bioecological model of human
development was originally created for analyzing the early stages of human
development, including childhood and adolescence. Therefore, generalizing it for
global, distributed software development can be questioned. However, it can be
assumed that as a software developer enters an OSS project, she encounters a
profound cultural adaptation process which requires learning both the shared
language of the community, its ways of working, and its prevailing socio-cultural
value systems.</p>
      <p>Our proposal is based on literature introduced in Section 1, along with
personal expertise of the authors. As three practitioners have given generous
feedback at di erent stages of its development, we claim that our application provides
a su ciently reliable representation of the software ecosystem context. However,
we acknowledge several places for improvement based on this feedback. More
attention to the microsystem level could be placed by observing or interviewing
new developers who still need help in comprehending their environment. Also,
further inquiries of active community members who mentor new contributors to
the project could shed light on the di erent exosystem phenomena. Finally, the
mesosystem layer could be amended by studying the role of corporate processes
as a factor of the ecosystem, which, however is applicable only to the special
case of commercially orchestrated communities.</p>
      <p>With- or without these improvements, our application can be used for
further studies - as shown by the rich explorations of the Bronfenbrenner's model
in social sciences (documented in e.g. [BM06]). Learnings from this line of
research can provide con dence for building new experimental research designs
that address interrelationships, change and continuity in the hybrid OSS
ecosystem context. While most activities in OSS development leave a digital trace,
researchers should be able to validate their results through rigorous practice.
Future studies could ask bold questions to nd e.g. connections in between
management and practice: "Do strict and explicit conventions lead to less testing
of newcomers' code contributions?", "Does a hierarchical governance model
decrease the need for coordination?" or "Does transparency of long-term
planning lead to prolonged commitment to the project?" As a more distant goal,
the veri ed model could help discovering metrics which operationalize health of
communities in terms of their intake and retainment of contributors. This could
provide new practical understanding on which factors should be placed in the
limelight of community management while designing entry paths for new
developers and facilitating their personal validation as new, productive members of a
community.
[ACG10]
[SSR02]
[TRGB15]
[WJ07]</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          [FBACF17]
          <string-name>
            <given-names>Oscar</given-names>
            <surname>Franco-Bedoya</surname>
          </string-name>
          , David Ameller,
          <string-name>
            <given-names>Dolors</given-names>
            <surname>Costal</surname>
          </string-name>
          , and
          <string-name>
            <given-names>Xavier</given-names>
            <surname>Franch</surname>
          </string-name>
          .
          <article-title>Open source software ecosystems: A systematic mapping</article-title>
          .
          <source>Information and Software Technology</source>
          ,
          <volume>91</volume>
          :
          <fpage>160</fpage>
          {
          <fpage>185</fpage>
          ,
          <year>2017</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          [FGMB14]
          <article-title>Fabian Fagerholm, Alejandro S Guinea, Jurgen Munch, and Jay Borenstein. The Role of Mentoring and Project Characteristics for Onboarding in Open Source Software Projects</article-title>
          .
          <source>In Proceedings of the 8th ACM/IEEE International Symposium on Empirical Software Engineering and Measurement</source>
          ,
          <source>ESEM '14</source>
          , pages
          <fpage>55</fpage>
          :1|-
          <lpage>55</lpage>
          :
          <fpage>10</fpage>
          , New York, NY, USA,
          <year>2014</year>
          . ACM.
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          [FSBM14]
          <string-name>
            <given-names>Fabian</given-names>
            <surname>Fagerholm</surname>
          </string-name>
          , Alejandro Sanchez Guinea, Jay Borenstein, and
          <article-title>Jurgen Munch. Onboarding in Open Source Projects</article-title>
          .
          <source>IEEE Software</source>
          ,
          <volume>31</volume>
          (
          <issue>6</issue>
          ):
          <volume>54</volume>
          {
          <fpage>61</fpage>
          ,
          <year>2014</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          <string-name>
            <surname>[GBICMR13] J. M. Gonzalez-Barahona</surname>
            ,
            <given-names>D.</given-names>
          </string-name>
          <string-name>
            <surname>Izquierdo-Cortazar</surname>
            , S. Ma ulli, and
            <given-names>G.</given-names>
          </string-name>
          <string-name>
            <surname>Robles</surname>
          </string-name>
          .
          <article-title>Understanding how companies interact with free software communities</article-title>
          .
          <source>IEEE Software</source>
          ,
          <volume>30</volume>
          (
          <issue>5</issue>
          ):
          <volume>38</volume>
          {
          <fpage>45</fpage>
          ,
          <string-name>
            <surname>Sept</surname>
          </string-name>
          <year>2013</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          [Jan14]
          <article-title>Slinger Jansen. Measuring the health of open source software ecosystems: Beyond the scope of project health</article-title>
          .
          <source>Information and Software Technology</source>
          ,
          <volume>56</volume>
          (
          <issue>11</issue>
          ):
          <volume>1508</volume>
          {
          <fpage>1519</fpage>
          ,
          <year>2014</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          [Joh06]
          <string-name>
            <given-names>Gary</given-names>
            <surname>Johns</surname>
          </string-name>
          .
          <article-title>The essential impact of context on organizational behavior</article-title>
          .
          <source>Academy of management review</source>
          ,
          <volume>31</volume>
          (
          <issue>2</issue>
          ):
          <volume>386</volume>
          {
          <fpage>408</fpage>
          ,
          <year>2006</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          [Kil14]
          <string-name>
            <given-names>Terhi</given-names>
            <surname>Kilamo</surname>
          </string-name>
          .
          <source>Essential Properties of Open Development Communities. Doctoral dissertation</source>
          , Tampere University of Technology,
          <year>2014</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref8">
        <mixed-citation>
          [KM06]
          <article-title>Mik Kersten and Gail C Murphy. Using task context to improve programmer productivity</article-title>
          .
          <source>In Proceedings of the 14th ACM SIGSOFT international symposium on Foundations of software engineering</source>
          , pages
          <volume>1</volume>
          {
          <fpage>11</fpage>
          . ACM,
          <year>2006</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref9">
        <mixed-citation>
          <string-name>
            <surname>[LW03] Karim R Lakhani and Robert G Wolf. Why Hackers Do What They Do: Understanding Motivation</surname>
          </string-name>
          and E ort in Free/Open Source Software Projects. Social Science Research Network,
          <volume>49</volume>
          :1{
          <fpage>27</fpage>
          ,
          <year>2003</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref10">
        <mixed-citation>
          [MFH02]
          <string-name>
            <given-names>Audris</given-names>
            <surname>Mockus</surname>
          </string-name>
          ,
          <string-name>
            <surname>Roy T Fielding</surname>
            , and
            <given-names>James D</given-names>
          </string-name>
          <string-name>
            <surname>Herbsleb</surname>
          </string-name>
          .
          <article-title>Two case studies of open source software development: Apache and Mozilla</article-title>
          .
          <source>ACM Transactions on Software Engineering and Methodology (TOSEM)</source>
          ,
          <volume>11</volume>
          (
          <issue>3</issue>
          ):
          <volume>309</volume>
          {
          <fpage>346</fpage>
          ,
          <year>2002</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref11">
        <mixed-citation>
          [MH13]
          <article-title>Konstantinos Manikas and Klaus Marius Hansen. Reviewing the health of software ecosystems{a conceptual framework proposal</article-title>
          .
          <source>In Proceedings of the 5th International Workshop on Software Ecosystems (IWSECO)</source>
          , pages
          <fpage>33</fpage>
          {
          <fpage>44</fpage>
          ,
          <year>2013</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref12">
        <mixed-citation>
          [NYN+02]
          <string-name>
            <surname>Kumiyo</surname>
            <given-names>Nakakoji</given-names>
          </string-name>
          , Yasuhiro Yamamoto, Yoshiyuki Nishinaka, Kouichi Kishida, and
          <string-name>
            <given-names>Yunwen</given-names>
            <surname>Ye</surname>
          </string-name>
          .
          <article-title>Evolution patterns of open-source software systems and communities</article-title>
          .
          <source>In Proceedings of the international workshop on Principles of software evolution</source>
          , pages
          <volume>76</volume>
          {
          <fpage>85</fpage>
          . ACM,
          <year>2002</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref13">
        <mixed-citation>
          [SAW94]
          <string-name>
            <given-names>Bill</given-names>
            <surname>Schilit</surname>
          </string-name>
          , Norman Adams, and
          <string-name>
            <given-names>Roy</given-names>
            <surname>Want</surname>
          </string-name>
          .
          <article-title>Context-aware computing applications</article-title>
          .
          <source>In Mobile Computing Systems and Applications</source>
          ,
          <year>1994</year>
          .
          <article-title>WMCSA 1994</article-title>
          . First Workshop on, pages
          <volume>85</volume>
          {
          <fpage>90</fpage>
          . IEEE,
          <year>1994</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref14">
        <mixed-citation>
          <string-name>
            <surname>[Sha06] Sonali</surname>
            <given-names>K.</given-names>
          </string-name>
          <string-name>
            <surname>Shah</surname>
          </string-name>
          . Motivation,
          <article-title>Governance, and the Viability of Hybrid Forms in Open Source Software Development</article-title>
          .
          <source>Management Science</source>
          ,
          <volume>52</volume>
          (
          <issue>7</issue>
          ):
          <volume>1000</volume>
          {
          <fpage>1014</fpage>
          ,
          <year>2006</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref15">
        <mixed-citation>
          [SSGR15]
          <string-name>
            <given-names>Igor</given-names>
            <surname>Steinmacher</surname>
          </string-name>
          , Marco Aurelio Graciotto Silva, Marco Aurelio Gerosa, and David F Redmiles.
          <article-title>A systematic literature review on the barriers faced by newcomers to open source software projects</article-title>
          .
          <source>Information and Software Technology</source>
          ,
          <volume>59</volume>
          :
          <fpage>67</fpage>
          {
          <fpage>85</fpage>
          ,
          <year>2015</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref16">
        <mixed-citation>
          [SWvK15] [Wel11]
          <string-name>
            <given-names>Srinarayan</given-names>
            <surname>Sharma</surname>
          </string-name>
          , Vijayan Sugumaran, and
          <string-name>
            <given-names>Balaji</given-names>
            <surname>Rajagopalan</surname>
          </string-name>
          .
          <article-title>A framework for creating hybrid-open source software communities</article-title>
          .
          <source>Information Systems Journal</source>
          ,
          <volume>12</volume>
          (
          <issue>1</issue>
          ):7{
          <fpage>25</fpage>
          ,
          <year>2002</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref17">
        <mixed-citation>
          <string-name>
            <given-names>Mario</given-names>
            <surname>Schaarschmidt</surname>
          </string-name>
          , Gianfranco Walsh, and
          <string-name>
            <surname>Harald F O von</surname>
          </string-name>
          <article-title>Kortz eisch. How do rms in uence open source software communities? A framework and empirical analysis of di erent governance modes</article-title>
          .
          <source>Information and Organization</source>
          ,
          <volume>25</volume>
          (
          <issue>2</issue>
          ):
          <volume>99</volume>
          {
          <fpage>114</fpage>
          ,
          <year>2015</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref18">
        <mixed-citation>
          <article-title>Lessons learned from applying social network analysis on an industrial Free/Libre/Open Source Software ecosystem</article-title>
          .
          <source>Journal of Internet Services and Applications</source>
          ,
          <volume>6</volume>
          (
          <issue>1</issue>
          ):
          <fpage>14</fpage>
          ,
          <year>2015</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref19">
        <mixed-citation>
          <string-name>
            <given-names>Friederike</given-names>
            <surname>Welter</surname>
          </string-name>
          .
          <article-title>Contextualizing entrepreneurshipconceptual challenges and ways forward</article-title>
          .
          <source>Entrepreneurship Theory and Practice</source>
          ,
          <volume>35</volume>
          (
          <issue>1</issue>
          ):
          <volume>165</volume>
          {
          <fpage>184</fpage>
          ,
          <year>2011</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref20">
        <mixed-citation>
          <string-name>
            <given-names>Donald</given-names>
            <surname>Wynn</surname>
          </string-name>
          <article-title>Jr. Assessing the health of an open source ecosystem</article-title>
          .
          <source>In Emerging Free and Open Source Software Practices</source>
          , pages
          <volume>238</volume>
          {
          <fpage>258</fpage>
          . IGI Global,
          <year>2007</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref21">
        <mixed-citation>
          <string-name>
            <given-names>Joel</given-names>
            <surname>West and Siobhan O'mahony.</surname>
          </string-name>
          <article-title>The Role of Participation Architecture in Growing Sponsored Open Source Communities</article-title>
          .
          <source>Industry &amp; Innovation</source>
          ,
          <volume>15</volume>
          (
          <issue>2</issue>
          ):
          <volume>145</volume>
          {
          <fpage>168</fpage>
          ,
          <year>2008</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref22">
        <mixed-citation>
          <string-name>
            <given-names>Yunwen</given-names>
            <surname>Ye</surname>
          </string-name>
          and
          <string-name>
            <given-names>Kouichi</given-names>
            <surname>Kishida</surname>
          </string-name>
          .
          <article-title>Toward an understanding of the motivation of open source software developers</article-title>
          .
          <source>In Software Engineering</source>
          ,
          <year>2003</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref23">
        <mixed-citation>
          <string-name>
            <surname>Proceedings</surname>
          </string-name>
          . 25th International Conference on, pages
          <volume>419</volume>
          {
          <fpage>429</fpage>
          . IEEE,
          <year>2003</year>
          .
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>