It is absurd model for evaluation of people that solves only one problem: how to make evaluation easy for the administrators and soft on researchers. We should rather find and implement a model that leads to best research results possible. I think one such model would be this: give tax money only to people whose results of their last work have proven to be beneficial to society (for other people outside their interest group, not only colleagues in the same field). Impose reasonable set of practical goals and supervision (roadmap) on people who want their work financed and periodically re-evaluate. Leave basic, speculative and other research of unclear significance to be financed privately, by patrons, no tax-based financial support.