Last week, I attended my first Academic Publishing in Europe (APE) meeting. Celebrating its 10th anniversary, the meeting is held each year in Berlin. Its attendance is capped at 200. It was a very useful meeting to attend for a number of reasons. The attendees came from across Europe primarily, with many from the Dutch and German publishing houses, providing a mix that we rarely see in the US or even in the UK. The meeting itself struck me as more willing to court controversy, which was helpful because this generated insights and better discussions during breaks and social events. There was a lot to talk about also because European policymakers are much more adamant about open access (OA), which is speeding up its implementation and revealing interesting stresses and contradictions.

Because of this, I found myself having thoughts I haven’t had before, at least not so clearly.

“The #1 issue in public access is the public funding of science.” This quote — which came from Fred Dylla as he discussed how US federal research funding as dropped as a percentage of GDP from more than 2% in 1965 to about 0.7% in 2013 — started a major cascade of thinking, especially after learning that the European science initiative, Horizons 2020, is allocating funding to support OA APCs while cutting funding for basic research overall. In the US, funding in constant dollars for research has declined more rapidly over the past few years. Wellcome Trust has set aside millions to support eLife, funds that may have otherwise gone to research grants. I think all the papers published in eLife would have been published elsewhere anyhow. In the UK, millions of pounds have been diverted from research budgets to support OA APCs. Overall, this is creating a situation in which more research is being published OA, but less research is being funded in general. The value proposition of scientific research is widely understood, while the financial benefits of OA are speculative at best. Even if they exist, if there are fewer scientists pursuing research agendas, then who is using all this pre-paid information? The Horizons 2020 trade-off is especially blatant. There is clearly an opportunity cost to OA funding as it is currently realized, and this is something that should be addressed quickly so that we don’t rob Peter (research dollars) to pay Paul (APCs). So while OA may be an issue some wish to advocate for, the larger issue right in front of us is that research budgets in major economies are shrinking, leaving many questions unaddressed and fewer scientists working full-time.

Is there a coming “baby bust”? As China has ramped up its research outputs and polished its abilities in both research and research reporting, we’ve seen another nation with as many researchers as the United States emerge. Along with increased pressures to publish overall in established markets, China’s rise has created a glut of papers on the market. Publishers are scrambling to deal with through new journals, mega-journals, and so forth. Plans are being made for more journals to cope with the rising number of papers. However, there may be a demographic time bomb in our midst stemming from the funding problems mentioned above. In our plans, either consciously or subconsciously, we may be straight-lining research outputs based on the rapid increase we’ve seen over the last decade. But with grant recipients growing older and young scientists unable to find stable funding and work, we may be creating the conditions for a “baby bust,” a generation with fewer researchers overall, which will lead to a drop in research outputs. Or perhaps with Western funders diverting money to OA APCs instead of adding money, we will only see the “baby bust” in Western research programs, giving China — which has a more strategic and committed government overall — a clear advantage.

Is the six-month embargo going to unite subscription and OA publishers? European policymakers are making noise about a six-month embargo rather than the more common 12-month embargo. However, with Gold OA proliferating rapidly in Europe and elsewhere, this embargo shift has the potential to threaten OA APCs as much as it may threaten subscriptions — that is, why pay so much for a few months of access? Either model finds itself facing this question if the embargo shrinks. Continued moves in this direction may create an uncommon alignment between publishers using the Gold OA business model and those using the subscription business model. And when a publisher uses both, they may be doubly motivated.

Creative Commons has important flaws, as does its implementation. Last year, I dug into Creative Commons and found it lacking for a variety of reasons. It seems I’m not alone. If you’ve ever bothered to read a Creative Commons license — something few of the publishers in attendance have actually done, it turns out — you begin to see how poorly constructed and weak the licenses are overall. A strange brew of copyright and contract law, CC licenses are untested, and two lawyers who presented each suggested that CC licenses may not pass muster in courts someday. For instance, some provisions in CC licenses are bound to be superseded by copyright law in most jurisdictions. Elements of CC licenses not informed or superseded by copyright law then depend on contract law to be binding. But as contracts, CC licenses are problematic, as there is no clear offer or acceptance, and one party — the authors — have no idea who they are contracting with. As for CC BY, the contract is probably de facto meaningless as no party is incentivized to enforce it. The CC mix of copyright and contract law (maybe that’s what “CC” really stands for — copyright + contract) may prove unworkable. While this is something the courts will figure out over time, we need to be aware we may be promulgating licenses that could be found lacking later. More bothersome even than this is the fact that we are seeing again and again that authors do not prefer the CC BY license, yet it is demanded by many prominent funders and within many definitions of OA. Some of these same groups are claiming that publishing should be “by scientists, for scientists.” If this is the case, why isn’t repeated evidence that CC BY is not preferred acknowledged by allowing for a more diverse approach to OA licensing? Robert Kiley from Wellcome argued that CC BY should only be abandoned if it has been shown to “cause harm.” That’s not a market-serving standard, especially when we have abundant evidence that authors believe it harms them and their interests. Authors have indicated repeatedly that they do not prefer the CC BY license. Why can’t we agree with them?

We have a lot to learn about data publishing, and need to improve our vocabulary. It was clear from discussions about data publishing that we are making big strides in some ways, but not in others, especially in talking about differences and boundaries. In some cases, “data publishing” was defined as posting the underlying data; in other cases, it was expanded to include code and scripts involved in models and analyses. Where and when “code publishing” makes sense may be an entirely different matter from data publishing, and it may require a completely different set of approaches and policies. We probably shouldn’t conflate the two. Human subject data is categorically different, as well, yet we blend it into these discussions in a way that drags the discussions down. We should probably separate it out. Biological data has another set of concerns, as do materials data. And so forth. Refining our discussions will require really hard and disciplined work, as will creating the repositories once we’ve decided what is going to happen.

Are non-profits structurally less able to innovate? During one of the dinners associated with the meeting, I struck up a conversation with a scientist who also studies the psychology of innovation. It was a fascinating discussion. One of the traps that can derail innovation is a lack of follow-through — that is, ideas start, but then the will to execute them, overcome unexpected barriers, or invest sufficiently to realize the goal falters. It is difficult for commercial organizations to innovate, but often they have a consistency of leadership and a culture of long-term financial planning that help support consistent effort and innovation projects. Non-profits, on the other hand, churn elected leadership regularly. Many elected society leaders have one-year terms, and bring a completely new set of priorities to the organization when they arrive. This continuous need for management to reeducate and refocus draws resources and attention away from new initiatives, making them less likely to succeed. Ultimately, a lesson of “we can’t innovate” may take hold in an organization, shutting down innovation initiatives firmly. It’s something to think about.

There were a lot of great topics covered at the meeting — reproducibility, climate science data publishing, emerging data standards, and university presses included. The group was convivial and navigated difficult conversations with respect and genuine curiosity. We can only hope to emulate a similar tone in every venue.