1. Traditional peoples who depended on their local ecosystems for their essential needs, have accumulated by trial and error a rich body of local environmental knowledge, and in several cases elaborated resource management systems, and developed institutions appropriate for implementing these systems. Biodiversity conservation appears to be integral to many traditional management systems from tropical forests to coastal fisheries. For example, some groups manipulate the local landscape to augment its heterogeneity, use conservation 'rules of thumb' in their management practices, and integrate the production of multi-species systems. Thus, self-interest of traditional peoples has been key to biodiversity maintenance. As traditional peoples are integrated into the global economy and come under various pressures, they often lose their resource base, and in the long run, their knowledge systems, social institutions, and their world view which shapes their relations with the environment. The process of decoupling of traditional peoples from their resource base is likely to reduce the resilience of their social systems, as well as their local ecosystems through biodiversity loss. The two are related, and the reduction of resilience will make both social and ecological systems more fragile. One challenge for biodiversity conservation is to learn from the knowledge-practice-belief complex of traditional peoples. But a perhaps more important challenge is to implement elements of it in systems which allow feedbacks from the environment, and respond to them in a more resilient way than do current day management practices.
2. In the past many development programs have failed because the users were neglected. Their knowledge and practices, means of communication, value systems, and sociocultural background in general were seldom taken seriously. This attitude is changing in the face of reality. It seems to be an appropriate time for researchers and scientists to reevaluate their importance, and restructure their research and development models accordingly so that mistakes made in the past will not be repeated in the future.