How should science be evaluated? Does the possibility exist of using other criteria beyond bibliometric impacts, which value different facets of the social impact of research?