It is generally accepted that all research should begin with a ‘good’ research question, but ethnography rarely conforms to this norm. In fact Hammersley and Atkinson (1995) point out that the research question usually changes during the course of an ethnographic study. We go in wanting to know one thing and as our knowledge of the context deepens our focus often shifts; therefore it is more useful to begin with an interest than with a rigid question. Of course, walking into a field site with no guidelines can be overwhelming even for the most experienced ethnographer. The chaotic multitude of stimuli that assails you on entering a ‘new world’, and the my triad of potential interpretations of this stimuli, demand some guiding principles, however flexible they might be. It is important to accept that we need to have a ‘springboard’, but the purpose of that stating point is to guide the observation, not to restrict discovery.