The transition towards sustainable development inevitably implies a multiplication of the factors to be considered. The problem is particulary evident in the domain of energy systems for which methods allowing a holistic comparison between technologies options are still only embryonic. However technologies or combinations of technologies are not fixed but like natural creatures they adapt to their environment (essentially based on economical and regulatory factors so far). In our view it is essential to establish scientifically sound and coherent methodological approaches which would allow technologies or combinations of technologies to freely compete within the same framework of values. The internalization of pollution costs (external costs), real or fictive, is at present, one of the less bad solutions to reach this objective. One of the approaches developed at LENI and using this approach (one alternative of so-called environomic optimization) has been demonstrated in several cases of complex energy systems and some of the results are presented. This approach which uses powerful genetic algorithms helping to master complexity, offers good prospects to reconcile some of the wishes from stakeholders, from the decision-makers to the engineers. Such holistic approaches are needed to efficiently approach the ambitious objectives of the 2 kWyear/(year.inhabitant).