Software Engineering for the Next Internet

The Internet no longer meets the world´s requirements. Recognition of the Internet architecture´s inherent flaws has opened a window of opportunity for fresh thinking: granting agencies in both Europe and North America are funding large programs to lay a foundation for the next Internet. The skills and perspective of the networking community are necessary but not sufficient to meet the needs of those who develop and deploy networked applications. The current Internet exhibits problems due to prejudice against application servers, neglect of the functional aspects of networking, poor understanding of abstraction and composition, and conflation of concerns. These problems make some applications very difficult to build and deploy, and they do not appear in the networking research community´s mainstream agenda for improving the Internet. Software engineers are already solving some of these problems by means of middleware. It is time for software engineers to broaden and deepen their interest in the Internet, so that we can design a hierarchy of composable, optional, functional overlays as the architecture of the next Internet as seen by applications. At the same time, we must work with the networking community to merge our top-down viewpoint with their profound bottom-up understanding of network resources and performance. In addition to its other content, this paper is intended to provide a quick tutorial on the Internet for software engineers, and a good reading list.