Megajournals

The trend towards Open Access has catalysed the creation of many new journals and new publishers. BioMedCentral, established in 2000, was a pioneer of open access publishing, launching a large number of journals. Public Library of Science (PLoS) initially established a small number of high-level journals, then in 2006 it launched PLoS ONE. This was the first of a new kind of journal, later dubbed mega-journal. PLoS ONE aimed to publish any article that met the test of scientific rigour, and eschewed any measure of importance or impact in its editorial and peer review process. In 2010, PLoS ONE published 6,749 articles, making it the largest journal in the world (by volume). Its success helped to persuade the mainstream publishing industry that fee-paid open access was a viable business model.

Recently I invited representatives from a number of open access publishers to discuss megajournals. Five of them gave presentations to an audience of scientists here, and one visited me subsequently to inform me about their operations.

Nature Publishing Group (Scientific Reports)

NPG is a privately-owned business that publishes Nature and a range of other prestige titles: the Nature-branded titles such as Nature Genetics and Nature Medicine. Nature Communications was launched in 2010 as the first Nature-branded title with a paid-OA option. About 55% 43% of its articles are published as open access. It is not a mega-journal, but is aimed at a notch below the other Nature-branded titles.

Scientific Reports was launched in 2011 as a mega-journal. It publishes articles from all areas of natural science. It will publish any article that is technically sound in method and conclusions, with no need for conceptual advance, novelty or impact. Negative results are accepted and results with a narrow community of interest. There are strict submission guidelines for authors in terms of the article structure. An editorial board of 500, all practising scientists, and an advisory panel, manages the peer review process. Referees are sent a template for their responses, in order to achieve a faster review time. Some submissions are papers rejected by other NPG journals. Workflows are automated and thus scalable. The average time from submission to publication is 102 days. At present the acceptance rate is about 70% 55% – over 450 papers have been published since June 2011. The journal website promotes new papers on the home page and also uses download and social media metrics to show popular articles.  It is also possible to browse by broad topic. Press releases are issued for selected papers and they are promoted via Facebook and Twitter. The article processing charge (APC) is GBP £ 890 / US$ 1350.  Articles are published under the CC-BY-NC-ND or CC-BY-NC-SA licence, but authors can now choose CC-BY if they wish [added 11 Jul].

Public Library of Science (PLoS ONE)

PLoS is a not-for-profit publisher that was established by academics for academics. It publishes two highly-selective journals using internal editors, and four less selective journals with academic editors. PLoS ONE covers all fields of science, aiming to publish any paper that is ethical, has scientific rigour with conclusions supported by the data, and that is properly reported. The potential impact of the paper or the size of the audience are not important. Internal editors do some ethics checks on articles before sending for review. There is an academic editorial board with 3,200 members. The production process is streamlined with no copy editing or author proofing. The acceptance rate is 62% 65%. A proportion of papers rejected by PLoS ONE do get published elsewhere. Thus far i In 2012 the journal expects to publish about 25,000 papers. have been published. The journal website allows readers to add comments or notes to articles and also collates social media mentions, including Mendeley as well as citations from various sources. You can browse by subject and also see the most viewed articles by subject. The APC is US$ 1350.  Articles are published under the CC BY licence.

 BioMedCentral (BMC Series)

BioMedCentral is owned by Springer, one of the largest scientific publishers. BMC publish four flagship (more selective) journals, a data journal, the BMC series of about 60 journals, and a growing number of journals affiliated with other organisations. Strictly speaking BMC does not publish a megajournal, but it can be argued that the BMC series taken as a whole functions as a megajournal whilst retaining subject-specific identity. These journals aim to publish sound science and do not filter by interest. The series as a whole has well over 1000 2000 submissions per month, while the acceptance rate varies between individual titles from 45%-55%. There is an in-house executive editor but section editors and associate editors are all practising scientists. Interesting articles are highlighted on the BMC website and for some articles commentaries are commissioned and further promotion is carried out. Reader comments are also shown. The medical titles use open peer review, and the articles’ pre-publication history is shown. The biology titles do not have open peer review, though BMC Cancer does have some basic science articles with open peer review. Another peer-review innovation is the re-review opt-out. The APC for the BMC series is between GBP £675 (for Research Notes) and GBP £1445 (for the flagship journals). The standard article-processing charge for BMC subject journals is £1230.  Articles are published under the CC BY licence.

FEBS (FEBS Open Bio)

FEBS is a not-for-profit organisation: its affiliated societies have about 40,000 individual members across Europe. FEBS’ publishing profits help to support its other activities promoting the molecular life sciences. It publishes three selective journals and has just launched FEBS Open Bio, published on their behalf by Elsevier. The new journal is for papers in molecular and cellular life sciences in health and disease, covering basic, translational and clinical sciences. Currently most submissions are papers rejected by the other three FEBS journals, but direct submissions are also encouraged. The emphasis is on work that is technically sound and appropriately described. Though novelty is encouraged, sounds science is the key. The journal has a rapid review process and thus far an acceptance rate of 68%. There is minimal copy editing. The academic editorial board is currently just 16 but this is being expanded. The journal uses the Elsevier article of the future format and shows some article level metrics. The APC is Euros € 1200. FEBS currently holds the copyright of articles  but allows the equivalent of the CC-BY-NC-ND license.

 Company of Biologists (Biology Open)

CoB is a not for profit organisation. Income from publishing helps to support its other scientific programs. It publishes four selective journals and launched Biology Open in January this year. It covers all aspects of biological sciences. It publishes articles that are scientifically sound but reviewers are not asked to try to judge the impact of the work, which will be left to the community post-publication.. Thus far 75% of submissions have been cascaded from the other CoB titles. Out of 170 submissions it has published 62 articles. The review process is swift, with a first decision after nine days and from acceptance to publication is just four weeks. The editor-in-chief and editorial board members are senior academics. An editorial board of 30 is being formed. The APC is $1350. Articles are published under a CC-BY-NC-ND licence.

Frontiers 

Frontiers is a Swiss-based publisher that was established by scientists for scientists. It publishes 12 main journals, with two more on the way (actually it’s more complicated than that as, for historical reasons, neuroscience is broken down into a series of several separate journals). It is not a mega-journal and does not promote itself as such but taken as a whole the Frontiers in… series has some similarities to the other titles here. The review process emphasises accuracy and validity not significance. Strict guidelines for authors and a clear template for reviewers help to deliver a fast review time, three months from submission to acceptance. Reviewers first evaluate articles independently from each other, following a standardized review questionnaire. Then, authors and reviewers participate in an online discussion until a consensus is reached. Reviewers’ names are published on the accepted article. The editorial board is 25,000 strong and all editorial decisions are handled by scientific editors. About 450 articles per month are published currently. The website is quite complex, breaking down journals into specialist areas, and sometimes into research topics. This helps to give a community feel (each niche area has its own page) but can make it confusing at first. There is a system of tiers, which once again seems a good idea but in practice is a bit opaque. Tier 1 is research articles; tier 2 is review articles; and there are two further tiers not yet populated. The idea is that authors of important research articles, as judged post-publication, are invited to write commentaries or focused review articles. The site also has a research network where you can establish a profile.  This is integrated with the publishing workflow so you can readily see the status of articles you have submitted or are reviewing/editing. Frontiers seems to be attempting more than just creating another journal, but has re-engineered the whole publishing process. Articles on the website include a limited range of article level metrics and other links. The APC is Euros € 1600 but a system of discounts means that the average fee paid paid is substantially less than that. Articles are published under a CC-BY-NC licence.

General

It is instructive to compare these journals. They all share a number of features:

• Sound science
• Impact not required
• Academic editors
• Automated, scalable workflows
• Fast turnaround time
• APCs around GBP £ 1,000
• Post-publication promotion
• Article-level metrics

One or two still “encourage” novelty, but do not require it. How low is the bar on originality? If a previously-published finding seems to be in need of additional evidence then another paper that replicates it is worth publishing. Would a scientifically sound, but quite banal, replication of a long-established fact be published? Presumably not, which suggests that there must still be some subjective judgement being applied in these journals. In practice the publishers said that it has not been a problem – they have not had large numbers of banal submissions.

All emphasise the use of academic editors rather than internal editorial staff, though in most of them there is still a limited role for internal editors. All of them claim to have a fast turn-around time, and to have automated systems that are scalable. There is a question about the willingness of reviewers to comply with demands for rapid response. If asked to complete the process within 14 days, how many will just say “Sorry, too busy. Find another reviewer”?  Two of the journals are genuinely broad; two have multiple subject foci; two have a definite subject bias. I wonder whether the subject-focused and multiple subject-focused journals will have more success in the long run, if Cameron Neylon’s comment about the benefits of community focus in science publishing is correct:

Because PLoS ONE and other wide scope journals are covering wide areas of research, they necessarily need processes that can cover many data types and different disciplinary approaches, and these are still human, and therefore relatively expensive, processes. It is also fairly difficult at scale to place as much reliance as you might like on community peer pressure to contribute — and this may be a real advantage that society and independent journals have — a close knit community can effectively run shoestring budget journals.

The APCs charged are mostly similar, about GBP £900, though BMC and Frontiers are a little higher.  The fees are certainly well below the average figure mentioned by the Finch committee. All have waivers available in cases where funding is really not available.

Most of the publishers also publish more selective journals, and are ‘cascading’ papers rejected by those journals into the less selective megajournal. This seems to make good business sense.

Article-level metrics (ALM) and post-publication promotion are another common theme. ALM are still in their infancy and, to be honest, a bit underwhelming in most cases.  But when an article is heavily read/downloaded/mentioned then they come into their own.  Such articles will be promoted on the megajournal’s website, and/or elsewhere on the publisher’s website. Some of the megajournals also offer to issue press releases for key papers.  If this post-publication promotion turns out to be popular with authors it could become a key advantage for the journals that do it successfully. I am not sure whether size (big publisher, big website, high hit rate) or focus (dedicated user group, high levels of interest) will win out in that race for eyeballs.

The licence under which articles are published also varies. The two established OA publishers (BMC and PLoS) both offer pure CC-BY, which will soon be required for work funded by the Wellcome Trust and the UK Research Councils. The others offer only more restrictive licences.  I expect that conversations about this are taking place and it may change in future.

I should add that this post focuses on journals in biosciences broadly following the megajournal model.  There are a number of other new OA titles launched recently or about to launch that are following a more selective model – eLife, Open Biology, Cell Reports, mBio, G3. Basic facts and figures about these and other journals have been usefully collated by the library at UCB. Another new title that looks rather different has recently been announced by Peter Binfield, the founder of PLoS ONE.  Called PeerJ it will launch in the autumn and looks like it will drastically undercut all other OA journals. I wonder whether PeerJ will kickstart another series of imitators: just as we have got used to the idea of the megajournal so a new journal life form seems to be developing.

I expect that the list of pure OA journals will grow over the next couple of years and we won’t be asking “why are there so many new OA journals” but rather “why do we have all these old-style subscription journals?”.  An editorial last year in one of those old-style journals, pouring scorn on the OA journal model, suggests that some of them are getting scared.