Three long-standing questions still need to be addressed to stimulate innovation in the European Union (Nature 467, 1005; 2010).

First, to what extent can governments make informed choices about which areas should be stimulated by public (and private) funding of research and development (R&D)? Governments generally lean towards areas with a strong past performance rather than favouring those with a promising future. Are public agencies — or any other organization — capable of picking future winners?

Second, assuming that governments have the capability and remit to select promising areas, the next question is whether the European Union is the proper level for policy interventions. To put it another way: to what extent do European-wide innovation partnerships yield better products than national or regional ones? This everlasting debate becomes even more relevant in the implementation and feasibility of large-scale R&D projects. Perhaps one should accept a variety of spaces for public R&D intervention — some sectors require international research and innovation policies, whereas others are the realm of regional policies.

Third, there is the issue of how to organize innovation projects that address societal issues. You rightly point out the challenges of coordinating multiple-actor constellations. However, science and technology studies teach us that proactively involving stakeholders from different backgrounds and disciplines can be beneficial to the 'responsible' steering, utilization and implementation of R&D.