You mention the 'publish and be damned' model in academia, in which scientific work is disseminated, regardless of merit, for rapid and public online criticism rather than slow, private peer review (Nature 469, 286–287; 2011). For this to work, we must devise new ways to link widespread Internet discussions to and from an original paper.

Under current academic publishing models, one could easily miss substantive reactions to a paper that appear in other peer-reviewed journals. Even when a paper is retracted, research shows that this information is poorly disseminated and that the paper can continue to be cited widely and positively for years afterwards (K. M. Korpela Curr. Med. Res. Opin. 26, 843–847; 2010).

The growth of blogs, Twitter and free online access have caused a welcome explosion in scientific content. But this is atomized and interconnected by a hotchpotch of linking and referencing conventions. If we are going to harness its true value, we shall need dedicated librarians and information scientists to find ways of automating the process of linking content together again. That in itself would be a transgressive scientific innovation.