On the Adoption Dynamics of Internet Technologies: Models and Case Studies

Today, more than any time in history, our life-styles depend on networked systems, ranging from power grids to the Internet and social networks. From shopping online to attending a conference via P2P technologies, the Internet is changing the way we perform certain tasks, which incentivizes more users to join the network. This user population growth as well as higher demand for a better access to the Internet call for its expansion and development, and therefore, fuel the emergence of new Internet technologies. However, many such technologies fail to get adopted by their target user population due to various technical or socio-economical problems. Understanding these (adoption) problems and the factors that play a significant role in them, not only gives researchers a better insight into the dynamics of Internet technology adoption, but also provides them with enhanced guidelines for designing new Internet technologies. The primary motivation of this thesis is, therefore, to provide researchers and network technology developers with an insight into what factors are responsible for, or at least correlated with, the success or failure of an Internet technology. We start by delving deeply into (arguably) the salient adoption problem the Internet has faced in its 40+ years of existence, and continues to face for at least a foreseeable future, namely, IPv6 adoption. The study is composed of an extensive measurement component, in addition to models that capture the roles of different Internet stakeholders in the adoption of IPv6. Then, we extend it to a broad set of Internet protocols, and investigate the factors that affect their adoptions. The findings show performance as the primary factor that not only affected the adoption of IPv6, but also plays a role in the adoption of any other network data plane protocol. Moreover, they show how backward compatibility as well as other factors can affect the adoption of various protocols. The study provides a number of models and methodologies that can be extended to other similar problems in various research areas, such as network technology adoption and design, two-sided markets, and network economics