Framing effects dictate logic by constraining it from the start. More graduate students will register by a set deadline if a late fee is imposed as a penalty than if it is positioned as a discount for early registration, even though the dollars are the same. Stating the dollar amount as a penalty (negative) increased compliance. It’s the same information, just cast in a slightly different way.
Over the past 10-15 years, our framework for debate in publishing has been about access — why someone should have access, who should provide it, how it gets funded, and so forth. Arguments to cast one as the positive state and the other as the negative state have raged for years. The debates have been polarizing and contentious and seemingly endless. But lately, it seems the framework has eroded, revealing another layer we need to examine.
Perhaps the access framework didn’t pose anything more than a business question. Perhaps a more fundamental framework has been emerging all along, one that poses both epistemological and existential questions.
David Worlock has been thinking along similar lines, I discovered after writing the bulk of this post. He thinks the debate has shifted:
Open Access, defined around the historic debate twixt Green and Gold , when Quixote Harnad tilted at publishers waving their arms like windmills, is most definitely over.
Worlock notes that speed is now a defining benefit of online publishing:
Dr Sam Gandy, an Alzheimer’s researcher, is quoted by F1000 as saying that his paper was published in 32 hours, and they point out that 35% of their articles take less than 4 days from submission to publication. As I prepare to stop writing this and press “publish ” to instantly release it, I cannot fail to note that immediacy may be just as important as anything else for some researchers – and their readers.
Instead of access issues, I think what we’ve begun arguing about has been about how best to publish, what to publish, and when to publish — thereby positioning convenience for the benefit of authors against selectivity for the benefit of a community.
Convenience is not the same as speed, however, as coverage of the Professional and Scholarly Publishing (PSP) meeting in DC earlier this year shows, notably the section describing when eLife bragged it would publish articles in 60 days, only to be upbraided by JAMA‘s Editor-in-Chief stating that JAMA publishes most articles in under 50 days. The difference lost in the exchange is that these journals are fundamentally different not because one is quicker to publish, but because eLife was designed to have a much higher acceptance rate. JAMA and other major journals publish quickly for a minority of authors, but they are not convenient for most authors. They reject too many articles to be convenient.
Convenience-based publishers promise fast, low-friction, and respectable publication, but make no promises about reaching the right community. Author needs are their primary focus. The “convenience” argument asserts:
- Most papers get published in any case, so why have an inefficient system of thousands of journals and multiple rejections prior to publication, rather than a single source and single process?
- Peer review is slow and capricious when it can be quick and direct
- Validation of results is all we need, and measures of novelty and importance are passé
- Costs are too great because there are too many layers and inefficiencies in the current system
- Publishers take advantage of inefficiencies to drive up prices and pad their publishing portfolios
- We need massive processing and publication systems because science is burgeoning
- Scientists should be able to publish when they want to publish
Community-based journals are more stringent, because they have in mind a specific audience and level of expertise within that audience. A significant amount of their value rests on their ability to fulfill audience expectations, which are much more important than author needs. The “community” argument asserts:
- Readers don’t have much time, and want a filter that increases their chances of finding articles that are relevant to them in the short time they have
- Peer review works best when the authors and editors understand who the research is intended for
- Novelty and importance matter to readers in a particular field, and differentiate journals naturally
- Getting the right content to the right readers increases publishing effectiveness
- Costs are increasing because review, editing, and publishing are becoming more complex
- There are more journals because there are more communities in science
- Scientists don’t have time to wade through a pool of undifferentiated papers
Some OA publishers have community-based journals and some subscription-based publishers now have convenience-based journals using the authors-pay model, confirming that this framework is not an access framework, but something else.
The new convenience movement has created mega-journals, which generally have higher acceptance rates and fewer iterations of review. So far, PLoS ONE is winning this game, with 90% of the mega-journal market under its control. It is the best convenience brand.
Pre-print repositories like arXiv are convenience-based experiences catering to active researchers. They don’t threaten the journals system, it appears, a fact driven home by the manner in which publishers are launching pre-print services themselves. Naturally, convenience-based publishers like PeerJ have pre-print options for their authors.
Author convenience has driven changes to journals over the past decade, with most publishing online ahead of print; establishing fast-track programs; performing more editorial rejection to send authors with incompatible papers onward earlier; and working to make time-to-publication shrink across the board.
But perhaps in catering to authors, we have lost track of the power and importance of community-based elements, and don’t tout these to authors routinely.
Returning to JAMA, it’s interesting to note that its appeal to authors consists of two parts convenience and one part community, promising first “Prompt Decisions, Rapid Publication Timelines,” then “Impact and Reach,” and finally “Author Focus.” Nowhere does it speak to the quality of the audience, only the size. But compared to one of their main competitors, the New England Journal of Medicine, JAMA does an admirable job. NEJM doesn’t mention its audience or make an appeal for papers, merely presenting an interface to submit papers to prospective authors. Other major community-based journals I looked are just as scant when it comes to communicating the value of the community of readers they have managed to generate over the years. Their author messaging is mostly about the mechanics of submission, review, and publication. Perhaps the one that takes the cake for minimalism is PNAS’.
A journal’s connection to its audience is one of its most valuable assets, yet this connection is not being used to draw forth the best papers.
Perhaps our brands are powerful proxies for community-based assumptions, but we shouldn’t take this for granted. After all, the industry-wide focus on convenience might be influencing your brand, making it more of a commodity of convenience. It might be good to revisit your instructions for authors to see if you’re touting how highly engaged you are with their community.
Why is this debate important? How we think about journals influences your policy approaches. Funders of all stripes (private foundations and governments) prefer convenience-based publishing, so much so that three of them started their own journal, eLife. They want rapid dissemination of their funded research, and convenience-based journals match their incentives on the supply side of science.
PubMed Central wants us to think about journals as undifferentiated inputs into a governmental repository, and seeks to obscure the branding and community-based differences inside an approach that makes publishing look more like a convenience-based approach. Simple. Vanilla. Just papers in a database. This can make it appear that mandates can be constructed easily and deployed uniformly. But domain and community differences are muted if not entirely erased in convenience-based approaches. A hard look across publisher sites would generate a different signal for policymakers, notably how varied the scientific disciplines and subdisciplines (and sub-subdisciplines) are, how editorial approaches vary between community-based journals, and how differently services are arrayed to meet audience needs.
How you think about these issues also colors how you think about usage data. I personally look askance at usage data coming from convenience-based journals because these journals have no way of validating their users as being qualified users in the journal’s target market. It’s unclear that any publishing goal is being met.
Does a difference in approach really amount to much of an editorial difference? You only need to look at a recent issue of a mega-journal to see the answer. In PeerJ recently, the content was a smorgasbord of unrelated topics — botany, biology, and one particularly amusing study of apostrophe use in eponymous diseases. It’s difficult to know who this journal is for. But it goes deeper. How can a journal with a range of papers like this ensure that the disparate manuscripts they are receiving get adequate peer review? Can one journal rigorously review a paper about microRNA and a paper about water capacity in soil in Korea and a paper about genomic sequences of sediments from the Red Sea and a paper about transcranial Doppler ultrasound and a paper about apostrophes in eponymous diseases, all equally well? And if these journals are conveniences for authors, what are the incentives for rigorous review and possible rejection?
Publishing is a service industry, but a key question for any service industry is, Who do we serve? Access models don’t answer this question. PNAS tends to serve authors, yet it is a subscription journal. PLoS Medicine serves a community, yet it is an OA journal.
No matter the access model, at their heart journals balance service to both authors and readers. Convenience-based journals primarily serve authors, and have acknowledged they do a poor job of attracting the right readers. Community-based journals primarily serve readers, but make life difficult for a majority of authors through rejection and revisions. There is likely no such thing as editorial equipoise in our realm.
Perhaps recognizing this subterranean debate will help us address the question we’ve been unable to address robustly in the Internet age — Who do we serve? More importantly, it may help us identify how best to serve them. “More articles” and “more access” are not adequate answers to this question.