What is the difference between classical remote sensing and sensor networks? What kind of data models that one can assume in the context of sensor networks? Can the sensors in the network concurrently contribute to the sensing objective, without creating network conflicts? It is becoming apparent that methodologies designed to resolve network resource allocation conflicts in the communications among open systems have several bottlenecks when applied to sustain networking among concurrent sensing nodes. Can we structure the network activities so that they are always directly beneficial to the sensing task? The goal of this paper is to articulate these questions and indicate how some resource allocation conflicts can be removed embracing collaborative networking approaches among the sensors.