Imagine a world in which the academic paper is no longer the ultimate goal, the inviolable end product of science. It’s hard, isn’t it? For centuries, scientific journals have offered a means to communicate new findings and ideas to the broadest possible audience — taking advantage of developments in the printing press and later, related developments to disseminate scientific knowledge to scholars the world over. Indeed, they have come to represent a sort of reification of this knowledge, as if the words and figures on their well-thumbed pages are a kind of Platonic ideal. But now all of that is changing. In “Beyond the paper,” an essay in a recent Nature special issue on the future of publishing, Jason Priem describes a widespread move towards “a more diverse set of outputs”: Researchers are sharing data in repositories such as GenBank, Dryad and figshare; moving beyond the conventional article format to include blog posts, interactive graphics and video; and taking scientific discussion to Twitter and other social media platforms. All of this can have profound implications for the dissemination of scholarly knowledge, which has for so long relied on a single, somewhat exalted product: the journal article. “The article was an attempt to freeze some part of the scholarly process for display,” Priem said. “The Web opens the workshop windows to disseminate scholarship as it happens, erasing the artificial distinction between process and product.” So what does this mean for the scientific journal? Will it whither away amidst the immediacy and expanded scope of these various other outputs? Phill Jones doesn’t think it will. Jones is a former biophysicist and editorial director and current VP for business development at ReadCube. Consider, he said, the role of the traditional journal: It offers quality control through editorial and peer review; provides an official version of record for science and academia generally; and facilitates communication of ideas, results and interpretations in a highly moderated way. “I believe that that process is still vitally important for science,” he said, “and despite its faults some version of it will have to continue into the future.” What we’re likely to see is publishers adapting to this new landscape, developing new tools and technologies to help researchers communicate their work within the same basic framework. At the same time, new companies are moving into the academic publishing space and, to varying extents, cleaving to the traditional journal model. figshare enables researchers to publish datasets and other outputs in “an easily citable, sharable and discoverable manner” (“it has been called a more useful version of supplementary information,” Jones said) and has partnered with Nature Publishing Group to launch the open-access journal Scientific Data.JoVE started out as a journal that purely published video and later started adding text for indexing purposes. In his Nature essay, Priem outlines the various ways in which traditional journals and the peer review system in place today might be swept aside by the “aggregated judgment of expert communities,” through personalized recommendation engines based on altmetrics and qualitative badges, for example. This may very well come to pass. Someday. For now, it seems that the journal — the idea of the journal as an authoritative repository of scholarly knowledge — is going to stick around.