The Model Evaluation Consortium for Climate Assessment (MECCA) from 1990–95 was an international consortium involving industry partners as well as research institutions in the dual attempt to advance basic science and to impact public and industrial policy-making. Led by the Electric Power Research Institute (EPRI) and the University Corporation for Atmospheric Research (UCAR), MECCA sponsored many numerical experiments on a CRAY-YMP supercomputer dedicated to climate modeling and produced information material with particular emphasis on assessing uncertainties involved in climate modeling for a range of potential users of climate predictions. As it turned out, it was difficult to align the goals of the climate modeling community with those of the climate impacts assessment community. Modelers primarily wanted to advance state-of-the-art General Circulation Models (GCMs). Impact analysts were interested in model evaluation in order to improve confidence levels in impacts predictions. An unconventional organizational scheme in climate science, MECCA also was a “social experiment” in bringing together diverse communities and overcoming mutual skepticism. Climate scientists were skeptical about the policy-making ambitions of MECCA and about industry’s motivation for getting involved in climate science, while industry doubted whether the scientists really made efforts to direct their work toward policy-relevant research questions. MECCA taught some scientists key lessons about the interactions between science and politics, but MECCA as such provided no effective answers to some of the basic skepticisms in the climate regime and consequently failed to reduce uncertainty on climate change issues.