Wednesday, December 02, 2020

Open Access: “Information wants to be free”?

(A print version of this eBook is available here)

Earlier this year I was invited to discuss with Georgia Institute of Technology librarian Fred Rascoe my eBook “Open access: Could defeat be snatched from the jaws of victory?” for Lost in the Stacks, the research library rock and roll show he hosts. 

Prior to the interview, Rascoe sent me a list of questions. As we did not have time to discuss them all during the interview, I decided to publish my answers on my blog. With the greater space available I also took the opportunity to expatiate at considerable length in doing so. This turned into another eBook!

Please note that what I say in the attached document is built on an interview. It is not intended to be any kind of prediction of the future; it is more an extended reflection after 20 years reporting on the OA movement, coupled with a heavy dose of speculation. Who knows, perhaps this will be the last thing I ever write on open access. Maybe this will prove my swan song.

I would also like to stress upfront that in the critique of the OA movement I make I don’t claim that my knowledge, or predictions, are superior to anyone else’s. This is just what I have concluded after many years observing the movement and reflects my current view on where I think we are today. It does also include a lot of factual data, as well as links and footnotes for those who like them. 

Importantly, while I do not consider myself to be an OA advocate, I admit that I was as naïve as anyone else about what the movement might be able to achieve.

Finally, while what I say might be slightly overweight in European developments, it may not matter if (as I believe is possible) events in Europe end up determining how open access develops globally. 

I say this because it seems possible that European OA initiatives will reconfigure the international scholarly communication system, and in ways that OA advocates will not be comfortable with. 

I would add that the main focus is on science publishing rather than HSS. 

The eBook can be downloaded here. (Health warning: it is 163 pages long). 

A short review of the eBook has been posted on Reddit here.


Wednesday, November 04, 2020

Community Action Publishing: Broadening the Pool

We are today seeing growing dissatisfaction with the pay-to-publish model for open access. As this requires authors (or their funders or institutions) to pay an article-processing charge every time they publish a paper it is felt to be discriminatory, especially for non-funded researchers and those based in the Global South  (see, for instance, here, here and here).

PLOS' Sara Rouhi

As a result, various alternative approaches are emerging intended to move away from APCs, including crowdfunding and membership schemes. Here institutions are asked to commit to paying an annual fee to a publisher, with the aim of pooling sufficient funds to cover the costs of making all the papers in a journal open access.

One of the more successful implementations of this model is the Open Library of Humanities (OLH), which has operated what it calls its Library Partnership Subsidies scheme since 2015.

Clearly, if the costs of the annual fee are to be viewed as reasonable by those asked to take part a sufficient number of institutions need to sign up.

The inherent weakness of the model is that a small group of community-minded institutions could end up paying all a publisher’s costs and everyone else would be able to “free ride”. In effect, those who join a membership scheme could end up shouldering the costs for everyone to have free access.

A key question is how many universities are willing to join an OA membership scheme. OLH co-founder Martin Eve has estimated that there are only around 300 libraries who will do so.

Nevertheless, we should not doubt that there is a real wish to move beyond APCs and many have come to believe that membership schemes are the best way of doing this. To be successful, however, they will need to broaden the pool of those willing to take part.

Annual Flat Fee

With this aim in mind, in October the Public Library of Science (PLOS) launched what it calls its Community Action Publishing (CAP) initiative for its two selective journals PLOS Biology and PLOS Medicine. The hope is that the journals can be gradually moved away from having to charge APCs and that the publishing costs can be shared as widely as possible.

Specifically, PLOS is inviting universities to pay an annual flat fee that will give their faculty unlimited publishing opportunities in the journals (there are separate “communities” for each journal) without the need to pay an APC each time. However, authors of institutions who do not join the scheme will be charged a non-member fee (NMF) that will increase in price over time, as below.

The rising cost of the NMF is intended to encourage universities who have not joined to do so in order to avoid their faculty having to pay to publish.

PLOS hopes that the annual fees will be low enough to attract a sufficient number of institutions but that the pooled funds will eventually be adequate to meet all the costs of publishing the journals. In the interim, the NMF means that there will be two separate revenue streams coming in and the free riding issue will be avoided.

The hope is that the NMF can be discontinued when the pilot period ends (it lasts from 2021 until 2023). However, says PLOS Director of Strategic Partnerships Sara Rouhi, that will be dependent on all the large institutions that PLOS is targeting joining the scheme. “If we don’t have a stable membership by then, we might have to offset lack of participation with more NMFs”, she told me. 

Tiers

The annual fee institutions will be asked to pay will be calculated by looking at their faculty’s publishing activity between 2014 and Q3 2019. Based on that they will then be assigned to a fee tier, as below.

To help broaden the pool, the annual fee will be calculated on the publishing activity not just of the institution’s corresponding authors, but of their contributing authors too. The aim is to ensure that “the cost of publishing is distributed more equitably among representative institutions.”

In other words, PLOS will calculate how often during the historical publishing activity period an institution was associated with both corresponding and contributing authors and then assign a weighting to each type of author. Specifically, contributing-author articles will be weighted at half that of corresponding authors.

As the PLOS FAQ explains: “Publishing activity is counted by determining the number of times an institution was associated with a corresponding author and the number of times an institution was associated with contributing authors. Papers where the institution is affiliated with the corresponding authors are weighted as 1 article and papers where the institution is affiliated with the contributing author are weighted as ½ article. (Multiple contributing authors from the same institutions are counted only once).”

Universities whose faculty have never published in the journals can also opt to pay the tier’s lowest fee in order to “insure” themselves against the possibility that one or more of their faculty might publish in the journal within the time period of the scheme (Jan 1st 2021 –  Dec 31st 2023).

In addition, says PLOS, Research4Life countries are automatically members of each community so researchers in those countries will never be subject to fees. And authors unable to pay non-member fees can apply for waivers as per the standard fee-waiver mechanisms offered by PLOS.

To signal its own commitment to the community PLOS has set targets for each journal and will cap its margin at 10%. Revenue exceeding the community targets will go back to members at renewal.

Time will tell how successful the scheme will be in broadening the pool beyond the 300 libraries who normally join OA membership schemes. Certainly, the challenge will be that much greater given that it is being launched at a time when many libraries are signing expensive transformative agreements with legacy publishers and universities are facing the financial impact of the pandemic. Some have also suggested that, as selective journals, PLOS Biology and PLOS Medicine do not publish a sufficient number of papers to make it worthwhile for many institutions to participate.

My initial thought was that PLOS will be targeting three separate types of institution (corresponding-author institutions, contributing-author institutions, and “insurance” institutions). Sara Rouhi suggests that this is not the way to view it. Below is an exchange I had with her.

Levers

RP: I understand the PLOS Community Action initiative will target 3 types of institution:


1.       Corresponding-author institutions.

 

2.       Contributing-author institutions.

3.        Insurance institutions (those institutions whose researchers don’t currently publish with the PLOS journals but who might want to insure against the possibility that one of their researchers will do in the future).

I guess there will be some overlap between 1 and 2 but I assume you will have done some calculations on the numbers of potential institutions for each category. If so, can you share them with me?

SR: While we initially grouped institutions this way, libraries and consortia consistently fed back to us that this was more complicated/confusing than it needed to be. So rather than think of different types of “institutions,” it’s easier to think of different publishing activity “types” – publishing as a lead author, publishing as a contributing author, or publishing with affiliations in both designations.

Given the selectivity and niche of these journals, some research-intensive universities publish little to not-at-all, and smaller institutions publish more frequently as lead authors. Simplifying the participation criterion to just publishing activity elides the traditional distinction of “research intensive” vs. “teaching” institutions etc.

I don’t want to add further confusion by grouping institutions as you’ve indicated. We purposely eliminated this distinction for clarity.

RP: Ok. I ask this question in the context of Martin Eve’s estimate that there are only 300 libraries that will support OA membership schemes. (I should add that he said this in the context of a new initiative from COPIM that aims [like PLOS] to “broaden the pool” of universities willing to join a membership scheme).

SR: As you say, Martin Eve identified about 300 libraries that will support OA membership schemes. There’s some nuance worth flagging, however. 

The TL;DR is that Martin’s statement is true if you assume institutions that are participating for largely altruistic reasons – because it’s the “right thing to do.” The PLOS CAP model couldn’t be predicated on that as a buying motivation. So it helps to go back to “collective action” basics. Credit to Dr. Kamran Naim for elucidating this at the Basel Sustainable Publishing Forum last week.

For collective action schemes to work there are generally two levers you can pull: 

 

1.       Encourage “pro-collective” behaviour by leveraging group affinity and social incentives

 

2.       Appeal to economic self-interest (aka “private benefit”)


From my perspective, SCOAP 3 and OLH lean heavily on the first lever. Participating in them is the “right” thing to do so libraries support them.

The challenge with only pulling on this lever is sustainability over time. SCOAP3 relies heavily on CERN’s ongoing support and has the challenge of large beneficiaries like Russia, Brazil and India not participating while benefitting from the open content.

OLH memberships are very small financial commitments that maintain OLH but make growth of the program difficult. Martin talked about this on a webinar we did for UKSG earlier in the year.

A collective model that focuses primarily on the second lever is Subscribe2Open. The primary benefit is getting the community to agree to make content open in exchange for a discount. If all members do not opt-in, the content stays behind a paywall. The combination of a private benefit (the discount) with the public good (making the content open) is a strong motivator for participation and a big part of why the model is so successful.

PLOS Community Action Publishing attempts to pull both levers. The pro-collective/group affinity/social incentives relates to the moral imperative – especially in a time of pandemic – to make biomedical research open to read and open to publish. Jeff Kosokoff’s comment that “Open Access is social justice,” speaks to this. Indeed, we have many commitments based on this mission-aligned priority for libraries. However what has interested partners from mere interest to actual commitment is the equitable fee structure (and relative affordability given the current budget crisis).

That said, without including appeals to economic self-interest, the PLOS CAP model would suffer from the same sustainability issues as other collectives. Hence our implementation of a secondary, “back up” revenue stream – non-member fees (NMFs) for authors from non-CAP members.

These NFMs are not meant to penalize authors but rather to encourage libraries to join rather than expose their authors to fees. For many institutions, the NMFs per article will be higher than the annual fee the library would pay to join one or both collectives.

Benefits

The NMFs help offset slow uptake of institutions who are not able to find the funds to support a model like this immediately. Thanks to those fees coming in, libraries can take more time to join and we can offer flexibility for institutions that need to leave the collective.

There are several benefits to this:

Enlarging the pool of institutions that would participate in this collective gives us the flexibility to create a much finer tuned fee structure with a long tail of low dollar fee tiers to accommodate institutions that do not publish frequently in either journal.

Those low fee tiers allow institutions strapped for funds and/or infrequent publishing institutions to participate because they support the moral imperative of equity and inclusivity that the model promotes. It simply doesn’t cost them that much to “do the right thing.”

Research intensive institutions then derive benefit from the “research light” institutions participating since they offset some of the cost burden. More institutions of varying publishing intensity means lower fees for everyone.

So, Martin is right if you’re counting only relatively wealthy institutions that want to do the “right thing.” If you start adding other considerations, especially private benefits (aka, we don’t want our authors to see non-member fees) and recognition of contributing author affiliations, you get a much larger pool and set of incentives to motivate ongoing participation.

Targets

As for our target institutions:

For PLOS Medicine there are 909 total institutions with some publishing history (some combination of lead and co-author affiliations) in the time period we evaluated.

44 of those are in Tiers 1-4

80 are in Tiers 5-7

The remainder are in Tiers 8-12

For PLOS Biology there are 1,557 total institutions with some publishing history (some combination of lead and co-author affiliations) in the time period we evaluated.

45 are in Tiers 1-5

152 are in Tiers 6-8

The remainder are in Tiers 9-12 

As I say, I wouldn’t necessarily correlate rough groupings of similar institutions as “corresponding author institutions, contributing author institutions, and insurance institutions” especially since it does not include the thousands of organisations with no publishing history who might want to participate. 

However, it’s probably fair to say that the highest tiers have institutions that published a lot in both designations. The middle tiers were a combination but at lower volume, and the lowest tiers were low volume and mostly in the contributing author designation.

Threat

RP: In a webinar you gave in September you said that if transformative agreements flourish there probably won’t be sufficient money left in library budgets to support schemes like PLOS Community Action. How big a threat do you think there is here?

SR: This, of course, comes down to how you define “transformative agreements.” If you mean the standard RAP/PAR deals that integrate historic subscriptions with open access publishing that stays revenue neutral, then yes. If these are negotiated with large commercial publishers first, we’re looking at new kind of “big deal” that locks in library monies with subscription publishers of hybrid open-access journals. Non-profits, small societies, and native-OA publishers may very well not make it out the other side of this transition (especially if “read” institutions’ subscription monies exit the system).

It’s hard to know how big the threat is as it ultimately comes down to choices libraries make. Many appear to be balancing both imperatives – looking at where they publish and spend the most and evaluating how mission aligned those outlets are. It’s not easy.

RP: So, what are your expectations about the choices that libraries will make with regard to the PLOS Collective Action Publishing scheme?

SR: It’s totally dependent on region, funding structures (block grants in UK versus how the US does it) and how radical libraries want to be about cancelation and re-negotiations.

I have personally been overwhelmed by the number of cash strapped organisations that are pushing to support this model and find the money.

The December commitment update will tell the tale!

RP: Thank you. And good luck!

 

 

 

Thursday, July 30, 2020

Unbundling the Big Deal: An interview with SUNY’s Shannon Pritting

The Big Deal has been a topic of heated discussion among librarians for some twenty or more years now. When first introduced, the attraction of the Big Deal was immediately obvious, since it allows a library to buy its faculty access to most, if not all, of a publisher’s journals at a much lower “cost per article” (discounted) rate. From the start, however, there were doubters.

Shannon Pritting

In 2001, the Director of Libraries at the University of Wisconsin, Madison, Kenneth Frazier, warned the library community of the dangers of signing big deals, or any comprehensive licensing agreement, with commercial publishers.

“The current generation of library directors is engaged in a dangerous ‘game’ in which short-term institutional benefits are achieved at the long-term expense of the academic community,” he warned, adding that big deals would weaken libraries’ ability to manage their journal collections, foist on them journals they “neither need nor want” and increase their dependence on publishers “who have already shown their determination to monopolize the information marketplace.”

Nevertheless, many libraries did sign big deals. And many later regretted it, not least because, having done so, they felt they had no choice but to keep renewing the contract, even as the cost kept going up and devoured more and more of their budget. 

Libraries felt trapped, conscious that if they did not renew they would have to go back to subscribing to individual journals at list price, which would mean being able to afford access to fewer journals, and fearful that when they discovered that journals they wanted were no longer available, faculty would revolt.

Over time, however, a greater willingness to think the unthinkable emerged, and some libraries began to cancel their big deals. And when they did so the sky did not fall in – which allowed other libraries to take heart.

The list maintained here suggests that libraries began cancelling their big deals as long ago as 2008, but the number doing so has been accelerating in the last few years. What has really focussed minds are the recent decisions by both the University of California and MIT to walk away from their negotiations with Elsevier rather than renew their big deals.

But it is not necessary to walk away completely in the way UC and MIT have done. Instead, libraries can “unbundle” their Big Deal by replacing the large package of several thousand journals they are subscribed to with a small à la carte bundle of a few hundred journals, and in the process save themselves a great deal of money.

What is helping libraries to make the decision to unbundle is the knowledge that more and more research is becoming available on an open access basis. In addition, new tools like Unsub are available to advise them on which journals they can cancel without too great an impact, and which journals are essential and so should be retained. 

Given the big savings that can be realised, and the pressure library budgets are under, unbundling is expected to grow, particularly in light of the straitened circumstances that libraries will find themselves in after the pandemic.

This year a number of US universities have unbundled in favour of smaller packages of journals, including UNC Chapel Hill, Iowa State University and the State University of New York (SUNY) -- a system of 64 institutions. Coming in the wake of UC’s decision to walk away from Elsevier these little deals have attracted a lot of attention. 

In Europe, by contrast, there is a greater focus right no on signing transformative agreements. In addition to providing reading rights, these new-style big deals include prepaid publishing rights to allow faculty to publish their articles on an open access basis. Amongst other things, these deals help assuage concerns about double dipping (where a university may end up paying both article-processing charges and subscriptions for the same journals). 

So how is a decision to unbundle made, and what are the issues and implications of making the decision? To get a clearer picture I spoke recently by email with Shannon Pritting, Shared Library Services Platform Project Director at SUNY. In April, SUNY replaced its Big Deal of 2,200 journals with Elsevier with a “little deal” of just 248 journals. By doing so, it says, it has saved about $7 million.

Unbundling raises a lot of questions, and I suspect we may not have answers to all of the questions for some time.

For instance, as more and more universities unbundle, how accurate will the calculations informing the decisions about which journals to give up and which to keep prove to be over time? This could have implications for, amongst other things, how much of the money that has been saved will need to be spent on obtaining paywalled articles through Interlibrary Loan (ILL) and document delivery services.

Moreover, since unbundling appears currently to be mainly a US thing (with Europe favouring transformative agreements) might we see a geographical divide emerge? If we do, what might the implications of this be, especially for the open access movement?

In addition, might unbundling encourage more researchers to use illegal services like Sci-Hub, and might unbundling see university libraries marginalised to some extent, especially if they do not play an active role in funding open access?

Please read on for the interview. 

Wednesday, February 19, 2020

PLOS CEO Alison Mudditt discusses new OA agreement with the University of California


The Public Library of Science (PLOS) and the University of California (UC) have today announced a two-year agreement designed to make it easier and more affordable for UC researchers to publish in the non-profit open-access publisher’s suite of seven journals.

Under the agreement – which is planned to go into effect this Spring – UC Libraries will automatically pay the first $1,000 of the article processing charge (APC) incurred when UC authors choose to publish in a PLOS journal.

Authors who do not have research funds available can request UC Libraries pay the full APC fee. The aim is to ensure that lack of research funds does not present a barrier for UC authors wishing to publish with PLOS.

The pilot is intended to test whether an institutional participation model that leverages multiple funding sources, rather than only grant funds, can provide a sustainable and inclusive path to full open access.

Below PLOS CEO Alison Mudditt discusses the new agreement and addresses some of the issues that the current trend for universities and consortia to sign so-called transformative agreements with legacy publishers raises for native open-access publishers like PLOS.


The interview begins …

 
Alison Mudditt
RP: The PLOS/UC agreement is essentially the same deal as UC signed with JMIR Publications in January. Is that correct?

AM: Essentially yes. UC has made their priorities for these agreements clear, so most UC deals will be very similar.

In addition, we are generating custom reporting for the UC to help them evaluate the efficacy of the pilot in bringing new authors to open access publishing while maintaining existing funding streams.

RP: Would I be right in thinking that these deals are native open-access publishers’ response to the transformative agreements that legacy publishers have been signing with universities and consortia like Project DEAL?

AM: While we can only speak on behalf of PLOS, this is certainly one of the drivers for us. We think that there is a significant opportunity for institutions and funders to prioritize partnerships with native OA publishers who stand fully aligned with their OA objectives.

We have been reassured by the commitment from institutions and consortia not to sideline negotiations with (and thereby disadvantage) native OA publishers.

RP: How many articles do you envisage UC faculty publishing with PLOS during the two-year period of the agreement?

AM: If we base it on previous years, then around 600-800 articles.

RP: How many articles a year do UC faculty currently publish with PLOS?

AM: Around 300 per year, across the seven journals.

RP: As I understand it, the agreement means that UC faculty will be able to publish in any PLOS journal and the first $1,000 will be paid by UC libraries. If the researcher has access to no research funds s/he can request full funding from the libraries. Is there any maximum sum agreed with UC libraries such that the funds could run out before the pilot ends?

AM: Yes, we have agreed to a capped total spend of $1.5M USD over the two-year period. This cap reflects library spend plus grant funding declared by authors.

Once that spending cap is reached, PLOS has committed to cover the cost of additional UC publications (assuming UC authors continue to faithfully declare their existing grant funding, a two year spend is most likely to fall between $1.2M and $1.25M).

$1.5M would demonstrate an unprecedented increase in publications from the UC – but of course, a key unknown is the level of demand once the barrier of APCs is removed.

If for some reason, we reach this cap earlier in the agreement period than expected (if at all), we have agreed to good faith renegotiations to ensure that both PLOS and the UC are protected from unanticipated surges in cost.

RP: Will the details of the agreement be published?

AM: Yes – as are all of the UC agreements.

RP: PLOS has an Institutional Account Program, of which I do not think UC is currently a participant. What is the difference between the agreement announced today and UC simply signing up to become a participant of the IAP?

AM: The Institutional Account Program is a simple direct billing program meant to minimize administrative overhead of APCs either through pay-as-you-go monthly invoices or debiting from a standalone account. While useful for mitigating administrative costs, it is not an OA deal, or transformative, in and of itself.

This new deal enables the UC to allow grants to cover APCs when they exist, so they can focus their support on where it is needed most (i.e. where authors do not have the grant funds). It introduces an organized, multiple-payer model of OA, which we think is important to test out.

And it meets our primary goals with new business models of ensuring that any author who wants to can publish with PLOS, regardless of ability to pay an APC.