The benefits you detail in your Editorial from collaboration between social and natural sciences (Nature 462, 825–826; 2009) can be extended to people like me. I develop software to support research and therefore need insight into how the research community functions.

The molecular biologists I work for are generous with their time in explaining their speciality. But to be useful, software must be compatible with the habits and attitudes of the community it serves. Most people participate in communities without being consciously aware of the norms they are following. I need to elicit the tacit knowledge from those I work for. Some of this is recorded by sociologists of science, and some is accessible if I can learn to use ethnographic methods.

The pace of change means that scientific communities regularly face new challenges requiring them to restructure themselves. For example, today's molecular biologists are gaining insight into the larger macromolecular machines of the cell. This progress depends on collaborations that combine techniques from different subdisciplines. To lead such restructuring, it is not sufficient to be recognized as a successful, ambitious researcher: an unusual level of insight into the functioning of the research community is needed as well.