e-Science (also known as cyberinfrastructure or e-research), has been driven by a variety of factors, including national policy initiatives, the needs of different scientific domains to cope with the 'data deluge', or needs for specific tools such as high-end visualization and data processing. This paper will present some of the research currently ongoing for the 'Oxford e-Social Science project: Ethical, Legal and Institutional Dynamics of e-Sciences'. It will also present an argument within the sociology of science and technology that suggests that knowledge production is increasingly being driven by research technologies. This argument will be illustrated with a number of examples from e-Science projects, placed in the larger context of research policy for e-Infrastructures, and a typology developed of the different kinds of research technologies and how their uses can be conceptualized.