In this report we have chosen to pursue a theoretical thinking, which has inspired and challenged our views of learning and development. It has been important for us to find some solutions to outstanding problems of learning. First of all, knowledge is anchored in a perception, which includes data and information. We have concluded that knowledge can not be managed or shared in a transfer sense. It is thus not possible to transfer knowledge. You can only transfer information. Two or more individuals are capable of sharing knowledge in only one sense. That is by developing knowledge jointly through some common activities. We adopt the concept of negotiation of meaning from Etienne Wenger to explain the process of knowledge sharing. One hurdle to overcome in this endeavour to grasp the fundamentals of knowledge management is the issue of tacit knowledge. It is difficult to complete carry the full load of implied knowledge across. In fact, it is impossible even if you attempt to make use of other means than verbal communication. A second step in the learning process is the anchoring in practice, which gives form and content to specific knowledge. The test of reality may strengthen or weaken a person’s knowledge. It does, however, also provide an opportunity for sharing of knowledge, even tacit knowledge, in the sense of developing knowledge. Through joint activities we can achieve a common understanding and obtain a shared repertoire – verbally and not least emotionally. This has led us to the concept of communities of practice, which refer to groups of people who share a concern, a set of problems, or a passion about a topic, and who deepen their knowledge and expertise in this area by interacting on an ongoing basis. Since communities of practice builds on personal engagement, the learning environment will be enriched, if this engagement is accepted and taken seriously. We have analysed the cooperation on the development of environmental information systems in Danish public institutions. Four dimensions have served as reference points for our identification of conditions to promote the formation and maintenance of communities of practice. The study is based on twelve semi-structured interviews with staff members of key environmental institutions. The analysis included two case studies on industrial environmental information system for the European Pollutant Emission Register (EPER) and digital mapping of regional environmental information. These two cases provide examples of collaboration in project groups and communities of practice. The environmental sector is undergoing massive reorganisation because of an administrative reform, which transfers significant environmental protection responsibilities from the regional level to local and national levels. In addition the public sector is undergoing a process of digitalisation. Higher demands on public access to information add to the complexity of the task. New skills in cooperation between public institutions and sectors are required. Our investigations do not provide a recipe for the organisation of future information systems, but identifies a number of factors, which we recommend in order to establish the right conditions for learning and development of information systems. The environmental information systems are central for environmental management, because it constitutes one of the main pillars of environmental planning, administration, permitting, inspection, public involvement and not least the development of knowledge on which the political decisions are made. Environmental information does not appear by itself, and since the field of environment is highly complex it is necessary to establish a solid cooperation between the authorities. This will require new forms of cooperation and new organisational structures. We are convinced that knowledge cannot be transferred, but it is possible to promote knowledge sharing through the creation of a minimal platform for the development of communities of practice. We find that Wenger’s analytical framework has helped us to identify strengths and weaknesses of existing cooperation. We do not claim that our approach sufficiently covers all aspects of learning environments, but our departure in Wenger’s social theory of learning has worked well.