For this issue, we asked three thought leaders for their opinions on the following:

Regarding Beall’s list of predatory publishers, what do you see as the ramifications of the proliferation of open-access publishers in business to collect available funder fees without doing appropriate peer review? How does such proliferation affect, for example, article quality, journal credibility, or tenure evaluations?

Here is what they said.

Rachel Burley Vice President and Director, Open Access, Wiley

Jeffrey Beall’s database of predatory openaccess (OA) publishers lists more than 300 publishers and journals—a 10-fold increase in just a year—and has attracted plenty of attention. It includes publishers that use deceptive practices, such as excessive e-mail spam to solicit manuscripts or reviewers, adding researchers to editorial boards without explicit permission, and hiding information about author fees.

The rapid growth of OA has created opportunities for new entrants to exploit the OA model. Setting up an OA journal is operationally less challenging than launching a subscription-model journal. Collecting multiple publication charges for individual papers is less complex than developing the infrastructure and operations required for the sales and support of content collections. And there is not the same need for a reliable flow of papers: an OA journal can in theory launch with a single article, because there is no requirement for a critical mass of content to justify a subscription fee.

As with any market, emerging opportunities attract new entrants ready to make a grab for market share. Startups typically emerge to serve the lower end of the market with the goal of building scale and ultimately competing with market incumbents. In scholarly publishing, predatory or unscrupulous publishers present a particular problem in that they can discredit the OA model at a time when the industry is moving to expand its uptake.

But predatory OA publishers also harm the markets that they claim to serve, creating more work for an already overburdened research and author community. For some authors, new OA policies may create additional work, for example, in verifying journal compliance, requesting funds for article-publication fees, and depositing papers in institutional or subject repositories. Add the effort required to vet an increasing number of requests to join editorial boards or review inappropriate manuscripts for questionable journals, and researchers will be spending more time than ever on the publication process rather than on the valuable process of research.

Credible publishers build their reputations on using ethical practices and stringent quality control and are fiercely protective of their reputations, which, in turn, ensure continued focus on ethical practices. That comes at a cost—viewed as too high by some in an “information wants to be free” era—but it does give authors the assurance that their research will appear in publications that use proper practices and that their names will be associated with trusted brands.

Beall has created a tool to help authors to identify journals to avoid. Ethical publishers will continue to apply rigor to their practices so that authors can be confident that they are putting their research and reputations in good hands when they choose to publish in these journals.

Michael Clarke Clarke Publishing Group

It is perhaps inevitable that the so-called gold open-access (OA) publishing model has led to predatory publishing schemes. To understand why, it is important to consider the costs inherent in developing a traditional, subscription-based publication. To launch a new subscription journal, a publisher typically incurs losses for 4–7 years while the publication slowly develops a base of subscribers sufficient to cover costs. The typical journal might break even in terms of its operating budget in that period, but it can be many years beyond this before the publisher recoups its initial investment. Launching a traditional subscription journal is a long-term investment that is not likely to produce a return to the publisher for 10 years or longer.

Contrast that high barrier to entry with that of a gold OA journal. The only fixed costs that an OA publisher must cover up front are those associated with staff salaries, an online hosting platform, marketing, and a peer-review system. Most other costs are variable: they are incurred only when a new article is published. They include costs of copyediting (if any), composition (if not automated), and XML markup (if full-text HTML is provided). Given the low fixed costs, a publisher can recoup its expenses much faster with an OA publishing model than it can with a subscription model. If a publisher is particularly successful in recruiting papers and in holding down costs, it is possible to recoup all investments and return a profit in the first year or two.

Given the attractive economics of starting an OA journal, more and more publishers are doing so. Because the barriers to entry are so low, the rush to OA is not limited to the established publishers. Many new entrants have launched journals in recent years. Some, such as eLife and PeerJ, have been launched by well-known agencies and investors with experienced staff who have long histories in scholarly publishing. Others have been launched by less reputable people whose sole purpose is to make a quick buck and who have no intention of providing the review and infrastructure (such as DOI deposit, legacy archiving, and abstracting and indexing) that is expected of scholarly publishers.

That second category of publishers is taking advantage of the increased volume of scholarly output that must find a home and of the proliferation of gold OA publications, many of which are reputable. Many OA journals are of recent vintage, so they do not always have impact factors or other indicators of quality. A disreputable publisher can launch a family of journals that have titles similar to those of more reputable publishers, list well-known scholars on their mastheads without informing them, and send out invitations to other scholars to submit their work. By the time the scholarly community figures out the ruse, the publisher can simply close up shop, bank its proceeds, rename its company, and repeat.

Subscription publishers do not have a monopoly on integrity (see Robert Maxwell), but the barriers to entry for this model are such that a publisher will see a return only if a publication maintains a good reputation and meets relevant performance metrics over a decade or longer. And although there are many reputable OA publishers that are producing highquality publications, the barrier to entry is low enough for less reputable organizations to flourish—at least for a time. As the gold rush slows and additional performance metrics are brought to bear (as well as such resources as Beall’s list), the claim jumpers will be easier to separate from those who have more legitimate interests.

Kent R Anderson Chief Executive Officer and Publisher, JBJS, Inc

The predatory open-access (OA) publishers that Beall identifies reveal to me what can happen when there are lower barriers to entry in scientific publishing. To reduce the barriers to entry, many OA publishers have redefined peer review. In addition, the OA business model eliminates the need to build expensive and complicated user-facing systems. Both approaches make it easier for predatory publishers to get into the market.

The term peer review has been appropriated by some OA publishers as though what they practice is equivalent to full peer review. But what they practice often has fewer safeguards and less rigorous practices than full peer review. Eliminating editors-in-chief is one way that mega-journals have lowered the bar for themselves. Senior editors and outside peer reviewers validate that material is appropriate for a journal’s audience and sufficiently important to merit attention. In many cases, fewer than half the manuscripts received at a journal that has strong peer review are sent to outside reviewers. Other processes can be sidestepped: for instance, checking for conflicts of interest, running antiplagiarism software, and securing author attestation forms. Predatory publishers have been shown to eschew even the more lenient practices, further lowering the barrier of entry.

It is also easier for predatory OA publishers to emerge because they do not have to build complex access-control systems and create and maintain customer databases and fulfillment systems. There are benefits to systems that capture information about users; scaling them up is integral to building a community. The systems and the data that they house are akin to the footprint of the community of readers and interested parties. Authors want to reach the relevant audience, and fulfillment systems and registration and access systems naturally gather such data. The OA model has no such capacity. One major megajournal has admitted that it does a poor job of getting its papers to a relevant audience. Because their journals are less technically demanding to launch, OA publishers can scale publishing platforms at a lower cost and faster because they do not have the barrier to entry that the above systems represent. That makes it possible for dozens of journals to be launched at once in the OA space, further encouraging predatory publishers.

It is not surprising that extreme practitioners of the OA model have emerged, especially given the lower barriers to entry that OA has created. The question for scientists and those who care about science is this: Are we better off having lower barriers to entry around information outlets that we all depend on?