Open access

“Open-access (OA) publishing is in vogue, and for very good reasons. Readers — especially those in developing nations — gain from the removal of subscription barriers, authors benefit from the additional exposure their work receives, public funders put tax-payers' money to better use, and publishers benefit from the economies of scale of an author-pays publishing model. However, such a model, in which costs are spread across authors instead of subscribers, is not well suited for highly selective journals with high rejection rates, for which readers vastly outnumber authors. Indeed, whereas high-volume journals such as PLoS ONE and Scientific Reports require a modest article processing charge of $1,350, selective journals with an OA option (hybrid journals) charge higher fees ($2,700 for Physical Review Letters and $5,000 for Nature Communications), and for highly selective journals such as Nature or the Nature research journals the amount would be much higher. Some publishing companies are therefore exploring additional revenue models based on providing added-value services for authors and readers. An alternative viable option is the publishing model pursued by the Public Library of Science (PLoS), in which the cost of running selective journals is subsidized by profit-generating, high-volume journals. Of course, the content of research articles is provided to publishers for free, peer-review work is voluntary and unpaid, and printed copies have long ceased to be an absolute need. It therefore makes increasingly more sense for journals (especially those belonging to high-volume publishers) to make access to published papers free, and charge authors for manuscript services (manuscript selection, peer-review management, editing, copyediting, typesetting, web hosting) and readers for premium services. Open-access publishing is thus transforming the scholarly publishing industry from being content sellers to becoming service providers. This transformation to OA publishing of scholarly content is happening quickly, driven by decreasing costs of digital versus print, by OA self-archiving mandates adopted by funders and institutions, and by public pressure into embracing the principles of open science. A study showed that in a random sample of 1,837 articles published in 2008 across multiple disciplines, 20.4% of the papers were freely accessible, 8.5% from the publishers' websites (so-called gold OA) and 11.9% elsewhere (green OA). But how quickly is the OA literature growing? According to recent data2, the projected yearly growth in the number of gold OA papers indexed by Thomson Reuters is 20%, whereas the total annual growth of published articles is 3.5%. One would thus expect that the output of OA journals is growing at a much faster pace than that of journals offering subscription-based access. However, as shown in Fig. 1, although early annual growth rates can be very high (approaching 100% for PLoS ONE and Nature Communications), the difference in sustained average growth rates between PLoS ONE and ACS Nano or Soft Matter — subscription-based journals with limited rejection rates — is only moderate. Of course, higher manuscript rejection rates can lead to slow output growth (Small) or inappreciable growth (this journal). This suggests that the expansion rate of OA literature will largely depend on the willingness of publishers to embrace economically viable OA publishing models3, and on their capability to innovate products and services that provide value to researchers. Also, at a projected 20% annual growth, by 2020 only about 27% of the papers published in that year will be gold OA. In the meantime, green OA should be adopted as widely as possible, for the benefit of researchers and the public. Many publishers, among them Nature Publishing Group (NPG), encourage authors to post their own version of the accepted paper (incorporating a reference and URL to the published version on the journal's website) in institutional repositories for public release six months after publication. Moreover, NPG and other publishers allow authors to post the originally submitted versions of manuscripts online (but not subsequent versions that evolve as a result of the editorial process), for instance on the ArXiv preprint archive. You are certainly welcome to do so."