Internet Success

Internet Success: A Study of Open-Source Software Commons

Charles M. Schweik
Robert C. English
Copyright Date: 2012
Published by: MIT Press
Pages: 368
  • Cite this Item
  • Book Info
    Internet Success
    Book Description:

    The use of open-source software (OSS)--readable software source code that can be copied, modified, and distributed freely--has expanded dramatically in recent years. The number of OSS projects hosted on (the largest hosting Web site for OSS), for example, grew from just over 100,000 in 2006 to more than 250,000 at the beginning of 2011. But why are some projects successful--that is, able to produce usable software and sustain ongoing development over time--while others are abandoned? In this book, the product of the first large-scale empirical study to look at social, technical, and institutional aspects of OSS, Charles Schweik and Robert English examine factors that lead to success in OSS projects and work toward a better understanding of Internet-based collaboration. Drawing on literature from many disciplines and using a theoretical framework developed for the study of environmental commons, Schweik and English examine stages of OSS development, presenting multivariate statistical models of success and abandonment. Schweik and English argue that analyzing the conditions of OSS successes may also inform Internet collaborations in fields beyond software engineering, particularly those that aim to solve complex technical, social, and political problems.

    eISBN: 978-0-262-30120-6
    Subjects: Technology

Table of Contents

  1. Front Matter
    (pp. i-vi)
  2. Table of Contents
    (pp. vii-viii)
  3. Acknowledgments
    (pp. ix-x)
  4. Preface
    (pp. xi-xii)
  5. I The Power of Openness and Internet-Based Collaboration
    • 1 The Importance of Open-Source Software Commons
      (pp. 3-12)

      In fall 2010, O’Reilly Media, Inc., and UBM TechWeb hosted the Government 2.0 Summit in Washington, DC (O’Reilly 2010). What was striking about this conference was the almost-constant underlying theme of the “power of openness” for innovation. The Government 2.0 Summit emphasized the idea that openness and innovation are important for the twenty-first century, but many might sensibly ask, What does openness have to do with innovation?

      When we reflect on history, the growth of the World Wide Web (WWW or the Web) from 1996 to 2000 provides one of the best examples of the power of this connection, because...

  6. II What We Know about Open-Source Software Commons:: Academic Research and Theory
    • 2 The Ecosystem
      (pp. 15-36)

      In chapter 1, we offered some powerful reasons for studying OSS. Here we provide a broader knowledge of the subject by distilling several years of study, starting with a bit of history that describes how OSS has changed over the last decade or two, and why we use theecosystemmetaphor to depict its current state. We then considercommonsmore carefully, because both the framework we use to organize our studies along with many of the factors we explore to understand the success and abandonment of OSS projects originate in commons literature. Our main objective in this chapter is...

    • 3 The Developer
      (pp. 37-56)

      In the first two or three years of the twenty-first century, economists were puzzled by the volunteer nature of OSS. Why would these developers be willing to give their intellectual property away? Why would they want to spend some of their free time to contribute to such projects? In this chapter, we study these individuals—the individuals who actually create OSS. We describe who they are and where they live, and explore what is known about why they participate in OSS projects. We also explain why the OSS developer is the key element in our research.

      To this end, we...

    • 4 Technological and Community Attributes
      (pp. 57-80)
      Meelis Kitsing

      During the first two years of our five-year research endeavor we studied literature, uncovering factors that might influence the success or abandonment of OSS commons, and developing hypotheses and research questions. At the same time, we were carefully considering how to define success and abandonment (see chapter 7). We were motivated by a strong desire to understand the quite-amazing OSS phenomena that we were watching evolve, and we continue to illuminate our early steps in that process of discovery in this chapter.

      In chapter 3, we used the term technological attributes to refer to a set of variables related to...

    • 5 Institutional Attributes
      (pp. 81-100)

      We have nearly reached the end of part II, where we have reviewed academic literature to find factors critical to the success or abandonment of OSS commons. Part II represents the beginning stage of our research endeavor, and many steps remain. In part III, following this chapter, we will find ways to define success and abandonment so we can classify existing OSS projects into those two categories; then we will find or generate data that provide measures for the factors we have presented as important. Finally, we will select statistical techniques that allow us to discover which factors are associated...

  7. III Our Empirical Studies of Open-Source Software Commons
    • 6 The OSGeo Case: An Example of the Evolving OSS Ecosystem
      (pp. 103-128)
      Meelis Kitsing

      OSGeo is a nonprofit foundation that provides support to a number of OSS projects creating software in the GIS domain. GIS software can be generally described as mapping software that links information to its position on the earth’s surface. For instance, we know one person who used GIS software to pinpoint the locations of nests of the national bird of the Caribbean nation of Saint Lucia in an effort to create a rich visual representation that would help the government preserve that species’ habitat. Open-source GIS software makes a major public goods contribution by giving average people, especially those in...

    • 7 Defining Open-Source Software Success and Abandonment
      (pp. 129-142)

      We now turn to an important topic that we only touched on in chapter 1: Howdowe define and measure success and abandonment in OSS commons?¹ Does success mean that a project has developed high-quality software or that the software is widely used? How might valuable software that is used by only a few people, such as software for charting parts of the human genome, fit into this definition? We wanted to discover what factors affect success or abandonment, but first we needed define and actually measure these two outcomes. This chapter tells the story of how we reached...

    • 8 What Can Data Alone Tell Us about Open-Source Software Commons?
      (pp. 143-178)
      Sandra Haire

      In this chapter, we are ready to use statistical methods to identify factors that influence OSS project success. We again turn to the 2006 data set of 107,747 SF projects that we used to construct our dependent variable in chapter 7, asking two important questions:

      1. Does this SF data set capture any of the technological, community, and institutional factors we described in chapters 2–5?

      2. If so, are any of these factors more often associated with successful or abandoned OSS projects?

      You may recall from chapter 7 that we see SF as a kind of remote sensor of OSS projects,...

    • 9 Filling Gaps in Our Data with the Survey on Free/Libre and Open-Source Success
      (pp. 179-200)

      We were fortunate to be able to conductThe Survey on Free/Libre and Open-Source Success, with the cooperation and active support of SourceForge Inc., the corporation that owns SF. This assistance made it possible for us to survey a random sample of all active SF developers, resulting in one of the few representative and potentially generalizable surveys existing in the OSS research field. We conducted the survey in fall 2009, after we had done our OSCeo case study (chapter 6), created our dependent variable (chapter 7), and completed our analysis of 2006 SF data (chapter 8). The work from all...

    • 10 Answering the Questions Raised in Part II
      (pp. 201-256)
      Meng-Shiou Shieh

      This chapter and the next one mark the pinnacle of our five-year effort to understand OSS commons. We closedThe Survey on Free/Libre and Open-Source Successat the end of 2009, and did the analytic work described in this chapter and the next during the fifth and final year of our research endeavor. This chapter is exciting because it provides information and insight about every hypothesis and research question that we posed in part II. In chapter 11, we will present statistical models that take all our variables into consideration at the same time—in order to find those that...

    • 11 Putting It All Together in Multivariate Models of Success and Abandonment
      (pp. 257-280)
      Meng-Shiou Shieh

      This chapter describes the final step in our empirical research, drawing on all the work presented thus far to try to answer our central research question:

      What factors lead some OSS commons to success and others to abandonment?

      We performed the statistical analysis below during the fifth and final year of our research effort. In this analysis, we put all the variables we have developed into the same metaphoric pot and use well-established statistical techniques to try to make sense of it all.

      At this point in our research we were still uncertain about the final outcome. Would we be...

    • 12 Thinking about Part III: A Review of Our Empirical Research
      (pp. 281-300)

      We undertook this study because the power of openness in combination with the Internet’s global reach excited us. As long ago as 2004, we understood that because of its phenomenal growth and impact, OSS might hold keys to understanding collaboration in the Internet era. Furthermore, studying OSS might help us recognize how these clues could be applied to solve problems in areas other than software. Thanks to a grant from the U.S. National Science Foundation, we were fortunate to be able to study OSS in substantial depth over the last five years.

      To begin our study, we identified theoretical factors...

  8. IV Conclusions
    • 13 Our Study in Perspective
      (pp. 303-316)

      Historically, much of the research on OSS has focused on large development team, high-profile projects. These kinds of studies are indeed important and useful, but too much concentration on these unusually large projects may skew our understanding of what is really happening in the OSS phenomenon. In much of this study—with the exception of chapter 6—we have tried to use data sets on OSS that are more closely aligned with the entire population of OSS projects. The great majority of these projects are made up of small developer teams interacting with an external user community. Our primary goal...

  9. Notes
    (pp. 317-322)
  10. References
    (pp. 323-342)
  11. Index
    (pp. 343-352)