new green section for report
Glenn Hampson
Hi Everyone,
In response to some of the feedback our Plan S report has received on Twitter, I’ve prepared a new page for this report that deals with the green open access ideas contained in this plan. In our original paper we didn’t comment on the green path allowed by Plan S. This wasn’t really an oversight because the main focus of Plan S is to banish subscription and hybrid publishing. But the critics are correct---it does also clearly stipulate that green is a compliant route.
So what might this mean?
I’m evidently grumpy this week so my first take (below) probably sounds harsher than it needs to sound---I’ll look at this again next week. In the meantime, your feedback/comments/criticisms/additions/tomatoes are welcome. My main concern at this juncture is making sure we’ve given this topic a complete look (not so much the copyediting, which Rob and Ilona will do). This page of text would be the new “concern #3” in this paper---starting on numbered page 7 right after the section on defining OA and before the section on going for the gold:
How GREEN how fast? Green open access is where research articles (as well as datasets, white papers, presentations, dissertations, conference proceedings and more) get deposited into "free to read" digital repositories. For most articles in most fields (with some notable exceptions) this usually happens in addition to publication in a traditional journals, not instead of.
There are a wide variety of repositories. Some are associated with journals, some with institutions; some host just pre-print versions of papers, others are massive warehouses that include versions of record with a range of copyright licenses, and still others even offer post-print peer review. Of all the spaces in scholarly publishing, green is probably the most varied, dynamic and fastest growing (note that figures 2A and 2B only count "self-archived" articles as being green, but in fact most green is "shadowed" by some other form of publishing). See Björk et al. 2014 for a more complete discussion of this form of open (it's a bit dated but still useful).
One observation, then, that has been raised about Plan S is whether this plan fully embraces green as an end point in the quest for open, or whether it is more a subset of other preferred outcomes (see MIT Libraries 2019). The plan’s implementation details specify three main routes to immediate compliance: (1) Publish now in a Plan S-compliant journal or repository; (2) Immediately upon publication, deposit the final published version in a Plan S compliant repository under a CC-BY license and without embargo; or (3) Publish CC-BY in a subscription journal that is covered by a transformative agreement.
Assume for the sake of argument that a rapid transition to green is being called for by point 1 and that authors will heed this call. What kinds of issues might arise?
· Disruption: What would happen if lots of researchers suddenly started posting all their work on Plan S compliant repositories? Certainly the pre-prints route is cheaper and faster, and is gradually increasing in popularity. But if we shift suddenly, what would this look like? Would this mean losing publishing capacity at universities, societies and journals? Losing filtering and signaling functions that journals currently provide? And where are we asking researchers to post? On bioarXiv or their university's in-house system? If the latter, what might this mean about potential fragmentation of the scholarly record? Caution here doesn't for a lack of enthusiasm, but before we start mandating a massive shift to green, we need to better grasp what the outcomes might be so we can ensure, at minimum, that research integrity is protected.
· Acceptance: Would publishing in green-only venues be an acceptable substitute for researchers? Not for all, but it would be for some (see the next bullet point). More study is needed to help prepare researchers and disciplines for this change in venue.
· Spotty demand: Green use varies widely by region (see table 2). Eger and Scheufen 2018 also note there is wide variation by discipline, with publishing practices falling into three distinct publishing cultures: (1) The gold open culture (generally the life sciences) with high use of OA journals but little use of repositories (gold open is discussed in the next section); (2) The green culture (generally physics, astronomy, math and business) with little use of OA journals but strong use of repositories; and (3) The grey culture (including social sciences, engineering, and chemistry) with mediocre use of both gold and green open access. There will be different levels of demand for a greener world (which of course begs the question of why should we push for a global shift to green instead of encouraging more discipline-specific shifts?).
· Interoperability and sustainability: As research repositories proliferate, so too do interoperability and sustainability issues. Standards have been proposed by Plan S, but these need by developed by repository experts; most of the repositories currently in existence don't comply with requirements set forth in Plan S. The Coalition of Open Access Repositories—COAR—has issue a detailed statement in this regard (COAR 2018). We should also not ignore the prospect building fewer and larger repositories (see the All Scholarship Repository discussion in OSIWG 2015), both to limit our interoperability and sustainability risks, and also to enhance the prospects of creating more value from truly vast databases of research knowledge.
The second compliance route—depositing articles in approved repositories after publishing them in approved journals—is not likely to boost the growth of green significantly. This is because Plan S-compliant green deposits must be non-embargoed, official versions of articles (versions of record or author approved manuscripts), licensed in CC-BY format. So, most authors who continue to publish in noncompliant journals will not be able to comply with Plan S by making their article free to read in a repository because this article may contain the wrong kind of copyright license, it may be delayed by embargo, and/or posting the official version of record may be disallowed. The only way to comply with green through the second route is to publish in a Plan S compliant journal.
So what does all this mean? It's impossible to say. Again, more study is needed. The incentive structure in academia is not going to change overnight, so most researchers in most fields will probably continue to seek out journal publishing opportunities—they won't skip journal publishing altogether. The risk of Plan S causing a massive, rapid shift to green is therefore unlikely. And assuming that most researchers accept the new publishing options created by Plan S (discussed in the next section), then green could marginally increase as a "shadow" output. Even if researchers choose noncompliant publishing options, green will still grow. So Plan S isn’t really doing anything to help green grow—a real and substantive shift to green will take different kinds of incentives and a lot more advance planning.
As always, thanks for your time an input.
Sincerely,
Glenn
Glenn Hampson
Executive Director
Science Communication Institute (SCI)
Program Director
Open Scholarship Initiative (OSI)
Rob Johnson
Hi Glenn,
Some initial thoughts on this...
INTRO SECTION
This all reads fine to me and provides a good overview of green.
OPTION 1 VS OPTION 2 vs OPTION 3
I think things then start to go awry a little, and you may be picking the wrong target by focussing your fire on point 1. I believe option/point 1 should actually be phrased as a 'Plan S compliant journal or platform' - not 'repository'. It also seems to me that you are setting up a straw man by suggesting later on that Plan S is encouraging researchers to 'skip journal publishing altogether', which I don't think is the case.
The implementation guidance states that 'Plan S applies to all scholarly output that is reviewed according to accepted standards within relevant disciplines', reaffirming the central importance of peer review. However, I read some of your text as suggesting that posting a pre-print (i.e. without peer review) would be compliant. Plan S encourages preprints, but it would only permit publishing in 'green-only' venues as a route to compliance where these also provide peer review. Most repositories do not.
For me, the platform option in Plan S and your option 1 is referring to services like Wellcome/Gates Open Research, F1000 Research, Science Open - these are not journals, per se, but do offer peer review, so are compliant with Plan S. Some of the concerns you raise are valid if publication on platforms grows at the expense of traditional journals, particularly those regarding filtering and signalling functions. That said, I don't think Plan S explicitly promotes or favours platforms over journals, it just permits both, so it can't really be criticised on these grounds.
ASSESSING THE MERITS OF GREEN OA
Your option 2 is what I see as the true 'green OA' option, where articles are deposited in a repository after publication in a traditional journal. A lot of the issues you currently raise under point 1, such as disruption, fragmentation, spotty demand and interoperability/sustainability, are legitimate objections to pursuing green OA, so should figure in an analysis of option 2. However, by focussing on these the OSI brief could just appear to be discrediting green as an option, having elsewhere argued that the 'gold-centric' approach in Plan S is flawed. If this is where we end up (gold won’t work, but neither will green), I don't see how this moves the debate on in any meaningful way.
I also think the latter part of the text ignores the possibility that journal publishers might change their policies in response to Plan S. Similar objections were raised about the lack of journals offering hybrid options when funders first began to require gold OA, but publishers actually moved pretty quickly to add a hybrid option when it clearly became in their interest to do so. Might we not see something similar for Plan S? For example, if 5% of a journal's submissions are subject to Plan S, wouldn't the easiest solution for all concerned be for the publisher to amend the journal's self-archiving policy to allow those authors (and only those authors) subject to the Plan to post a copy immediately, with a CC BY licence? The proportion of content that is then immediate OA is not significant enough to threaten the journal's subscription revenue, and this also allows authors to continue publishing in their venue of choice.
What might also be useful in this section is some reflections on lessons learned from the US public access programme, which is essentially a green mandate. If this is considered to have been successful, why is this? If not, why not? I suspect others on the list could contribute some useful insights here.
A SUGGESTED APPROACH
In summary, I would suggest that this part of the brief should acknowledge the limitations of green, most of which are already identified in your draft, and recognise that, in an ideal world, immediate, gold OA is preferable, and a worthy goal. I agree with many of your closing points on the need for more study and that Plan S as currently framed may not do much to stimulate green.
However, a further point you/OSI might wish to make explicitly is that in some disciplines, at least, existing scholarly culture and a lack of external funding are likely to make sustainable gold OA publishing very difficult to achieve. In these cases, the disruption entailed in a forced transition to gold might outweigh any benefit gained. Some form of green OA, potentially involving embargoes, could thus offer the best means of extending access without causing significant damage to scholarly communication processes within these disciplines.
I am sure others will have their own views, and I look forward to hearing them, but I hope this helps in refining the text a little further.
Best wishes,
Rob
Director
Research Consulting