Debrief – Amsterdam Call for Action on Open Science
PeerJ was fortunate enough to be invited to Amsterdam for the EU Presidency Conference on Open Science this past Monday and Tuesday (April 4-5, 2016). I was able to attend and represent PeerJ. We had some fantastic early Summer weather (save a spot of rain), which is always nice when traveling. Indeed, the weather seemed to put everyone in good spirits, for the optimism of a more open Europe was high.
Why this matters
The series of EU Open Science Conferences is an important event for not just everyone in the EU, but across the globe. The policies being drafted now and perhaps officially implemented by EU ministers in late May will set up precedents for all nations to look toward. And that affects every citizen. If we get it wrong, it could mean scientific progress isn’t as fast as we’d all like or could have achieved.
This debrief isn’t to go into every facet of the announced Open Science Policy, of which the Draft version was worked on at the conference and will be available for a week for further comment as I understand it. Rather I’d like to talk about some “offline” discussions that were happening behind the scenes.
EU Open Science Policy Platform
That said, to quickly summarize – the aim of this pan-EU effort is, as the name of the series of meetings suggest, to accelerate a transition to Open Science. This goes beyond just Open Access (OA), though the policy sets a goal of 100% Open Access compliance in the EU by 2020. Beyond OA, the policies aim to set into action a stimulus for opening up data outputs and resolving a myriad of Open Science policies across EU.
Briefly, there are five high-level action items, each with subitems that the Draft EU policy is looking to implement. I won’t list each subitem, but only ones that I’ll touch on further below:
- Removing barriers to open science
- Change reward and evaluation systems in science
- Create transparency on the costs of academic communication
- Developing research infrastructures
- Fostering and creating incentives for open science
- Mainstreaming and further promoting open science policies
- Stimulating and embedding open science in science and society
[Aside] Policy drafting and actionable outcomes
Having been to many conferences that aim to achieve “actionable outcomes” there are very few that actually have this happen. And so I was impressed that this meeting already came with a set of Draft recommendations for the EU Open Science policy. This, I view, is critical to create any sort of practical next steps and catalysis for change. The drawback to a “pre-draft” approach is that not all stakeholders feel as if they’ve been heard, and so how the revisions and comments are handled is equally critical.
— Dⓐniel Mietchen (@EvoMRI) 5 April 2016
Now, the offline discussions…
The coffee breaks were perhaps more interesting than the invited talks themselves. Although off the cuff and raw, the “offline” thoughts are usually the more accurate representation of what is happening in the environment, in this case, how close are we to achieving the goals of Open Science in the EU?
I highlighted three action items from the draft policy above, because although not formally connected in the draft, they were extremely intertwined in offline chatter (diagram below). Full disclosure, they are also connected to PeerJ; regardless, they would serve to enhance the entire market.
One of the most common items discussed was changing evaluation systems, i.e. losing the dependence on the Impact Factor. The Impact Factor is viewed by most as corrupting science (psst, review SF DORA). Changing the reward system may in fact be the linchpin in the diagram above. There are policies on the books to stop the secrecy behind publisher big deals in the EU and that is great, but getting rid of non-disclosure agreements will not lower prices by itself, which is the expected benefit of action item 4 of the draft EU policy, “Cost transparency.”
Why won’t ending big deal NDAs lower costs?
Cost is NOT the dominant factor when academics are deciding where to publish – prestige, i.e. Impact Factor, still rules. We already know that bundled subscription prices are obscene, and legacy publishers are increasingly bundling Open Access APC charges into these big deals. Transparency of the deals is of course needed, and may lower some costs, but that is unlikely to move the needle much. More proof of this is that Open Access APC pricing is fairly transparent, but what we’re actually seeing is prices going UP (and I’ll get back to that in a moment).
Creating new publishing models? Not without changing the reward system.
Much discussion was in excitement for new journals with better pricing, preprints, data citing, etc. The problem is that none of these have a leg to stand on, or at least it is a very expensive leg to acquire. New journals typically require hundreds of thousands/millions USD in capital investment as it takes years to build the prestige required for sustainability. Here, we may as well define “prestige” as synonymous with getting an Impact Factor. Indeed, perhaps an alternative to the high capital journal model are overlay journals. Overlay journals are too new to know for sure, will still require several years to obtain an Impact Factor, and are still dependent on the underlying infrastructure of a preprint service with expensive annual budgets. And of course preprints, data sharing portals, etc will never take off without changing the reward/evaluation system either.
PeerJ leaving money on the table is proof the system is broken
This is what I really want to get out there, low cost publishers like PeerJ are leaving money on the table. I heard this often at the conference in private discussion, and I know it is true. We’re definitely undercharging, particularly when you stack up the comparative value of journals.
There’s a terrible dynamic between obtaining an Impact Factor, the prices that are needed to be charged to survive long enough to get one, and then what can be charged once a journal does get the Impact Factor. This is not how you accelerate science, it’s the opposite of how to do it.
The way the current system is set up is to incentivize charging more, not less, for both Open Access and subscription journals. There are no dynamics that put downward pressure on pricing. Academic publishing is an inelastic market, that’s bad. The system is broken, but we can do something about it.
The only way to get an influx of innovation in the publication model, and actually lower costs, is to change academic incentives. We won’t see much traction until we can start evaluating research, not based on the journal it is published in (Impact Factor), but by its own merits. What those merits are and how to incent them is what needs to be critically examined in the EU Open Science policy platform drafts.
Jason Hoyt, PhD