OA Now Interview with Peter Binfield of PeerJ
Open Access Now 2013-06-19
In recognition of its first anniversary, the editors of OA now interviewed PeerJ Co-founder and publisher, Peter Binfield. Peter has worked in publishing for twenty years. Prior to Co-founding PeerJ, he served as the publisher of PLOS ONE, which he led to become the largest journal in the world.
1: First, could you survey the first year of PeerJ, noting some of the successes and challenges you have faced?
Certainly – this first year has been an amazing journey! We announced ourselves on June 12th 2012 with little more than a press release, the 2 co-founders, and a basic web site. But the response we saw as a result of that announcement was incredible – people were truly excited, and of course intrigued, about the potential for this model. In large part that was down to the new OA business model that we created for PeerJ (low cost lifetime memberships giving academics the rights to publish freely with us thereafter), however the names of the founders (Jason Hoyt – from Mendeley, myself – from PLOS ONE) and the financial involvement of Tim O’Reilly (of O’Reilly Media Group) also meant that people sat up and took notice. At that time, Nature described us as “a significant innovation” (http://www.nature.com/news/journal-offers-flat-fee-for-all-you-can-publish-1.10811); we were covered in places such as Ars Technica (http://arstechnica.com/science/2012/06/new-open-access-journal-aims-to-disrupt-scholarly-publishing/), Huffington Post (http://www.huffingtonpost.com/2012/06/12/peerj-oreilly-open-access-publishing_n_1589638.html) and the Times Higher Education (http://www.timeshighereducation.co.uk/420315.article); and Science magazine won the award for best headline with “New Open Access Journal Lets Scientists Publish ’til They Perish” (http://news.sciencemag.org/scienceinsider/2012/06/new-open-access-journal-lets-sci.html)!
We spent the next year building the company up to where it is today. That involved recruiting an Editorial Board of 800 Editors (including 5 Nobel Laureates); recruiting staff (we now have 5 staff total); building submission, peer review and publication software entirely from scratch; re-thinking the publication process from bottom up to work better in a ‘digital first Open Access’ mode; designing beautiful, user-friendly interfaces; and establishing ourselves with all the third party services and agencies who support the journal publishing ecosystem (for example memberships of OASPA (http://blog.peerj.com/post/50498436021/peerj-now-a-member-of-oaspa), COPE (http://blog.peerj.com/post/50412709909/peerj-member-of-cope) , and CrossRef; listings with the DOAJ (http://blog.peerj.com/post/50567500713/peerj-in-the-doaj), PubMed Central, PubMed (http://blog.peerj.com/post/49845635560/peerj-articles-in-pubmed-and-pmc) and Scopus (http://blog.peerj.com/post/47445954946/pubmed-central-pubmed-and-scopus-indexing-peerj); long term archiving at CLOCKSS and LOCKSS (http://blog.peerj.com/post/40018981867/archiving-and-peerj) and so on).
Having put all the pieces in place, we opened for PeerJ submissions on Dec 3rd 2012, published our first PeerJ articles on Feb 12th 2013, and launched PeerJ PrePrints (our preprint server) on April 3rd 2013 (http://blog.peerj.com/post/52783384704/celebrating-the-one-year-anniversary-of-peerj). Even in an established publishing company, with more resources and a pre-existing infrastructure, this would have been moving fast – and yet the whole ecosystem is now launched and running very smoothly. As of today, we have already published 89 PeerJ articles and 31 PeerJ PrePrints and several hundred articles are at some stage or other in our peer review system.
2: One of the most interesting aspects of PeerJ’s membership model is that one can imagine it developing and enhancing OA scholarly publishing communities. If this is indeed happening, could you describe the initial stages of community-based scholarship developing at PeerJ and how you anticipate this development to progress? If it is not happening, can you give some thoughts as to why not?
At the moment it is probably too early to provide concrete examples of this happening – we have ‘only’ published 89 articles and so the network and community effects that we expect to see don’t really have a big enough database to work on yet. However, I think you can see the beginnings of where this will take us.
I often tell people that because we have a membership model, and because we have a single system for our submission, peer review and publication processes then by natural consequence we have a very good understanding of how our users contribute to the overall system. As a result, as we go forwards, we expect to develop functionality which is more ‘person centric’, as opposed to ‘article centric’. This contrasts with a ‘normal’ publisher (either OA or subscription in fact) who typically doesn’t care or know very much about who their authors or reviewers are, or whether they have had prior interactions with them – these publishers have usually outsourced their peer review and/or publication systems to third parties and as a result their databases of users are split over several services and are not even disambiguated. Because of our membership model we have a greater knowledge of our users and this will then allow us to do new and exciting things with our functionality at the level of the individual user.
For example, you can already go to an individual’s profile page on PeerJ (https://peerj.com/MathewWedel/ ) and see interactions they have had in our system (for example as an author, editor, reviewer or commenter). And because we know what actions they have completed then we can award ‘contribution points’ (https://peerj.com/about/FAQ/academic-contribution/) to them, meaning that they can now be recognized for their involvement in the publication process.
Similarly, if an individual chooses to comment on a PeerJ article (as they can do on PeerJ PrePrints already) then the author, or other users, can choose to highlight the comment as insightful and ‘vote it up’ (see the Feedback on https://peerj.com/preprints/15/ for example). In this way, members will be able to demonstrate which areas they are expert in, and whether they have supplied insightful feedback.
And if you use our search engine (https://peerj.com/articles/) then you can see that we present faceted search results from article text, from preprint text, from user profiles, from editor biographies etc
Going forwards, expect to see us build more of these types of functionalities, which will then start to act on an ever widening database content and interactions.
3: Last year you emphasized PeerJ’s focus on the Biological and Medical sciences, claiming that “you need to concentrate on one thing at a time.” Over the past year, have you given any more thought to moving beyond these disciplines? In particular, what are your thoughts on how PeerJ’s model might work in disciples, in which published articles typically have fewer, if any, coauthors?
At the moment we have no plans to move outside these core areas. The disciplines we are publishing in have an output of well over 1 million articles per year, and these are also the areas that are seeing the most development and adoption of Open Access. Therefore, this is a market that we need to ‘crack’ first, before focusing elsewhere.
In addition, as you point out, there are fields with quite different publication profiles to ours – for example, humanities articles typically have just one author and can be very long in extent. Therefore, a move into any new area would need to be accompanied with an analysis of whether or not the business model works there. Clearly, anyone else is at liberty to emulate our model (and we have heard rumors of people considering this) and so it will be interesting to see if our model gets adopted and/or tweaked for other fields.
4: In response to the OSTP memo on public access to federally funded research, ARL proposed SHARE while a coalition of publishers proposed CHORUS. What role do you see OA publishers, such as PeerJ playing here?
Although I am certainly no expert in either proposal, my understanding is that both SHARE and CHORUS are, in large part, an attempt to fit a square peg into a round hole – i.e. one of their primary purposes would be to find a way to take ‘openly available’ (not necessarily open access) content published in a non-Open Access journal (together with ‘full’ OA content of course) and make it sufficiently discoverable and accessible that it would meet the requirements of the OSTP memo. As such, both schemes represent an attempt to provide a useful service during a time of transition towards ‘full’ Open Access.
Once we get to the point when all articles are ‘born’ OA (in an open access journal, not as part of a hybrid arrangement), with an appropriately liberal distribution license (we use CC-BY) then much of the problem they are both addressing will actually go away. The sooner we transition to full OA, and publish all content in the most openly available format, the sooner we can realize the benefits of OA that things like the OSTP memo are really aiming towards.
Therefore, the role of publishers such as PeerJ is to continue to push the world towards a situation where all content is published in as openly accessible a format as possible!
5: Could you explain the role you see download counts playing in evaluating the overall impact of a scholarly work? How does this apply to PeerJ PrePrints?
PeerJ advocates using alternate metrics (so called ‘alt-metrics’) to help evaluate scholarly work. The most ‘trusted’ metric is a citation from another scholarly work, however it is clear that there are many other possible indicators of impact, such as downloads, tweets, facebook likes, blog posts, news coverage, influencing governmental policy and so on. There are several groups now actively working on this problem and attempting to find as much altmetric data as possible and to discern a signal from the noise.
You asked about download counts in particular, and we would say that these are just one possible indicator of impact. In many ways simple download stats are really just a popularity rating and so it is important when considering them to know what they represent. For example, a more nuanced view of download counts would be to normalize them by subject area, and article age. And an even better approach would be to measure who it is that is downloading the articles, or how long they are spending looking at certain parts of an article etc. Therefore, although raw download counts are important (provided people recognize their limitations), they actually have the potential to be much more important in the future. Right now we are still in the early days of ‘alt metrics’ and there is a lot of work still to do!
As to how it applies to PeerJ PrePrints and PeerJ – for usage reporting, we currently provide number of visitors, number of page views and the hits from the various referral sources (something which few other publishers do). In addition, we provide a full suite of other metrics such as tweets, facebook likes, google +’s etc – all via Impact Story. We keep up to date with the latest developments in the alt metrics community, and as new metrics evolve (or new ways to evaluate the existing data are developed) then we will look to integrate those sources as well.
6: You recently announced that some institutions have entered into arrangements with PeerJ – can you explain how that works?
We have two ways that an institution (usually a university library) can work with us. Institutions can choose to ‘bulk purchase’ individual memberships for their authors in advance (and distribute them as they choose), or they can pre-pay for memberships which are then used by their faculty as and when they come to publish at PeerJ. Both options allow an institution to provide an Open Access option to a large number of their authors, in an extremely affordable way – literally for the price of just one year of access to one or two subscription journals an institution can provide hundreds of their faculty with a way to publish freely with PeerJ for the rest of their career. This has to be a great (and cost effective) way for institutions to help encourage their faculty towards publishing in Open Access as a matter of course.
So far institutions such as Duke, Birmingham, Nottingham, Newfoundland and Trinity have signed up for an arrangement like this and if anyone would like more information then they should direct their librarian to: https://peerj.com/pricing/institutions/