Technology Coupling and Diffusion of Innovation: Evidence from Open Source Software Development

Abstract

In contrast with the existing literature that focuses on the social process of innovation diffusion, this study asks the question: how the diffusion process of an innovation can be affected by its own technological attributes? We seek to address this question by focusing on how a technology is linked to its upstream and downstream complementary technologies. Based on the literature on product modularity and technology ecosystems, we argue that upstream technology coupling harms the diffusion of an innovation as it increases the technology’s performance uncertainty and exerts higher hurdles of learning for users. Meanwhile, stable coupling with one or a few major downstream complementary technologies not only stimulates diffusion, but also attenuates the negative effect of upstream linkage. The empirical analysis on the diffusion of open source computer program libraries, based on a unique data set of over 110,000 open source software programs and over 7 million usage by software developers, provides strong support to our hypotheses. This study contributes to the literature on technology diffusion by highlighting the critical role of technology itself in such process. It also advances the understanding of innovation ecosystems and competitive strategies based on open source technologies.