1 / 32

Cross country comparison on the factors determining OSS diffusion

Cross country comparison on the factors determining OSS diffusion. Spyridoula Lakka Teta Stamati Christos Michalakelis. Harokopeio University Athens. Open Source Software (OSS).

spyridon-galen
Download Presentation

Cross country comparison on the factors determining OSS diffusion

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Cross country comparison on the factors determining OSS diffusion Spyridoula Lakka Teta Stamati Christos Michalakelis Harokopeio University Athens

  2. Open Source Software (OSS) • Innovative model of software development, where software is open for inspection, modification and exchange without restrictions or discrimination. • High diffusion of OSS in different market segments: Linux, Android, Apache, Mozilla Firefox, Open Office, PhP, Java, MySQL, Moodle, Joomla, Apache OfBiz ...

  3. How is OSS diffused among countries? Limited Knowledge regarding: • OSS diffusion curves • Factors that shape OSS diffusion among countries

  4. Research Questions • How is OSS diffused among countries? • What theoretical foundation can be used to identify country level factors that affect OSS diffusion? • What are the factors that impact the diffusion process? • What is the impact of these factors at different stages of the diffusion process?

  5. Theoretical Framework Diffusion of Innovations Theory + Socioeconomic Theories • Endogenous Growth Theory • Exogenous Growth Theory • Institutional Theory

  6. The model • A country is conceptualized as a socio-economic system within which OSS growth occurs. • The model is based on the idea that the forces of growth to an economic system comprise of institutional, endogenous and exogenous factors : OSSit = F(Xendog, Xexog, Xinst)

  7. Endogenous growth theory • Economic growth is generated from within a system as a direct result of internal processes (Romer, 1986, 1994).

  8. Hypothesis H1: Telecommunications infrastructure has a positive impact on OSS growth. • Infrastructure such as the Internet and broadband connections are key elements not only for the diffusion but also for the existence OSS, as the OSS development model is totally based on virtual teams and remote management and collaboration.

  9. Hypothesis H2: Human capital skills and education can impact OSS diffusion. • Human capital is a key input to the development of knowledge, new ideas and products associated with technological progress. • Prior studies demonstrate that the average level of education and the quality of human capital are influential drivers for individual technology adoption. • Inexperienced and unskilled users would be very reluctant to migrate to an OSS system, due to the fear of lack of technological support and long term.

  10. Hypothesis H3: Innovation can be a ‘pushing’ factor for OSS diffusion. • The growth of OSS is mainly due to the large number of skilled and qualified developers with willingness to create and innovate. Thus, OSS is not only an innovative technology, but also is a continuous source of innovation. • Prior studies have identified the close relationship between OSS and innovation.

  11. Hypothesis H4. Economic growth is an impacting factor for OSS diffusion. • Previous research has showed that economic growth is directly related to technological growth. • Although OSS can be acquired at low or zero price, the study tests the possibility that economic development creates demand for more technologically advanced products such as OSS.

  12. Exogenous growth theory • Growth is primarily determined by external factors, such as the flow of goods, ideas, capital and technology innovations, rather than internal factors (Solow, 1965).

  13. Hypothesis H5: Technological openness can leverage OSS adoption among countries. • A country’s openness can be defined as the degree to which a country is open to business and economic influences through trade activities. It can be perceived as the external force that captures knowledge spillovers among countries . • The study considers ICT trade as the channel for achieving such spillovers .

  14. Institutional Theory • It attends to the deeper understanding of the processes by which social structures are maintained and reproduced (Scott, 2004).

  15. The quality and operation of institutions may have a severe impact on the way that social systems behave towards a technology. • Governments have the opportunity to intervene and take actions affecting the markets and the dissemination of new technologies. • Assuming that trade and innovation can affect OSS diffusion process (H3 and H5), the effects of government’s regulations and policies in these sectors are further examined.

  16. Hypothesis H6: regulation in trade could have a negative impact on OSS adoption among countries. • Hypothesis H7: regulation in IPR could have a negative impact on OSS adoption among countries.

  17. Hypotheses, constructs and measures

  18. DOI Theory • Describes the process of the diffusion of innovations through a social system, seeking explanations in terms of “how” innovations diffuse through it, by the means of appropriate mathematical models, the diffusion models.

  19. Bass diffusion model • The Bass diffusion model suggests that a new product adoption decision is driven by two factors: • coefficient of external influence, the influence that is independent of the existing number of adopters • coefficient of internal influence, the social influence of the existing number of adopters.

  20. Bass model limitations • Due to the use of time series, a small and a large country that have exactly the same number of adopters over time will have the same diffusion parameters. • The use of fixed time periods for all countries in a data set creates higher risk of left-hand truncation bias, where the estimates of the intercept of the diffusion curve are inflated for those countries that start their adoption process earlier than the time frame represented by the data.

  21. Left-hand truncation bias 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 1999

  22. Dekimp modified Bass model • ni,tis the number of adopters at time t, • Ni,t-1 is the cumulative number of adopters up to time t-1. • Si is the social system size, • Ci is the long-run penetration ceiling. • Bi is the diffusion growth rate. • Ai1 = ni,1/CiSi is the diffusion curve intercept.

  23. State-based diffusion parameters • Logistic models test covariates that affect the first year adoption (Ai1) and diffusion growth rates (Bi): • d1 and d2are vectors of parameters and Xiis the vector of the covariates for country i. • Xiis the vector of the covariates for country i.

  24. OSS diffusion model • Vector of covariates Xi Xi = (GDP, BBS, HCI, INNOV, ICTtrd, RTB, IPR) • Social system size Siis assumed each country’s population. • Long-term penetration ceiling Ci is defined as the proportion of the population aged between 15 and 65, as this age range should include the majority of OSS users. • Alignment of introduction timing of OSS: The model sets time t=1, the first year of adoption of each country, so that all countries’ diffusion will set off from a common point of the time axis.

  25. Dataset • The analysis is based in data from 25 countries over the period 2003-2008. • OSS penetration: cumulative number of subscribed users/developers per country in the SourceForge.net. Data queried out of the University of Notre Dame (UND) database which keeps SourceForge.net data for academic and scholarly research purposes. • Ci parameter: derived by the World Bank Indicators (population ages 15-64, % of total). • Si, parameter: extracted by the United Nations database

  26. Cross country comparisons Country listings according to: • Number of per country registered users in the SourceForge.net in 2009 (I) • Number of per country registered users, taking into account the parameter CiSi of each country in 2009 (II) • User rates the first year of adoption, Ai1 (III) • Diffusion Growth rate, Bi (IV)

  27. Results (NLS regression)

  28. Conclusions I • Factors that determine OSS diffusion: • Innovation • Quality of human capital • Telecommunications infrastructure • ICT trade • Regulation protecting IPR

  29. Conclusions II • OSS evolves in social systems where knowledge, education and creativity is highly developed. • Economic growth and less regulation barriers in trade affect OSS only in the beginning of the diffusion process. • Differences in the weight of impact at the two diffusion stages. Factors related to the quality of human capital and innovation are more critical in the growth phase of the diffusion. • Close relationship between OSS and innovation.

  30. Thank you for your attention.Any Questions?

More Related