This paper is an attempt to develop some fundamental aspects of Information Systems domain, which are generally hidden or misunderstood. Generally speaking an IS is generally considered only for its shaped objects : in the first hand, programs, methods, classes, systems, and in the second hand, interfaces, rules, processes, workflows… In fact, to produce so shaped objects, the IS development process needs to elaborate conceived objects. But the utility of these conceived objects are often limited in the private space of experts. Nevertheless this paper claims that their utility is much more important and their usability will become very crucial in the IS engineering. Furthermore and finally, the fancied objects are hidden in the most often cases : they appear only through objectives, that the IS should fulfil : so the IS development process would be only a problem solving process. However, many papers show that there is a deep gap between the management domain and the computer technical domain. Many of them try to build a bridge between these domains. This paper claims that the IS domain cannot be reduced to a simple bridge and the difficulties of the enterprises with their development with IS cannot be solved following this way. The only level where managers and computer specialists can cooperate with a minor risk of manipulation, appears to be the level of fancied objects. Then the overloaded complexity due to lazy positions in IS development disappears. The IS utility-usability becomes real stakes. Furthermore IS development process becomes a collective creative process, where IS becomes the means of expressing a collective sense to cope with complexity. Finally the paper illustrates this position through the example of the agile information systems (in particular with the M7 open source project).