OpenAlex is a database of academic authors, institutions and publications. Since its launch in January 2022, OpenAlex has received a lot of attention as an alternative to commercial research databases such as Web of Science or Scopus that would better meet academic needs and values. OpenAlex is based on a multitude of sources across all fields of science and languages, and on a global scale. A user can search by author, institution and research output, and select specifically by type of output (article, book, dataset, preprint, editorial, etc.), citations, publication date or availability in Open Access. The starting point for OpenAlex was the dataset of the discontinued Microsoft Academic Graph (which was the second largest academic search engine after Google Scholar), which was enriched and refined – a process that is still ongoing – to be able to be used as an alternative to commercial research databases for all kinds of searches and/or bibliometric analyses.

The OpenAlex data – which is shared under an open licence, namely Creative Commons Zero (CC0) – is available in three ways: via an online user interface (i.e. ‘OpenAlex Web’), via data snapshots (which enable you to save a copy of the OpenAlex database locally – as is at the time of download) and via the OpenAlex API. Use of OpenAlex Web, the data snapshots and the OpenAlex API is free of charge. There is a paid service which accomodates intensive use and offers additional support, but the free version suffices for the typical individual user.

The (lack of) cost, as well as the open philosophy behind it, is something that sets OpenAlex aside from commercial products like Web of Science and Scopus. These are expensive products and a recent study even shows that the companies behind these use specific sales strategies that maximise profits but come at the expense of the academic community. What is more, OpenAlex is lauded for its completeness and inclusivity. Web of Science and Scopus are selective databases, based on a curated set of sources (which has been criticized in the past for being too focused on particular disciplines, as well as specific languages, regions and publication types); whereas OpenAlex tries to be as complete as possible and is therefore not only more representative for disciplines like humanities, but also for the state of research in various languages on a global scale.

Quite a large number of studies analysing the quality and (dis)advantages of OpenAlex have been produced recently. The status quaestionis is:

• If one wants to get as complete a picture as possible of the research output of an author or of an institution as a whole (all scientific disciplines, all languages, all publication types), it is advisable to use OpenAlex.
• If one wants to map the OA availability of research output, it is advisable to use OpenAlex.
• For specific bibliometric analyses, it may be advisable to use Web of Science or Scopus due to the selectivity of the database and the (for the time being at least) relative superiority of the metadata, provided that one is aware of the limitations (e.g. in terms of scientific discipline, publication type and language).
• When compiling systematic reviews, it depends on the exact objective. If one wants to map scholarly literature on a particular topic as completely as possible, it is advisable to use OpenAlex; if, on the other hand, one wants to obtain a selection of scholarly literature that is representative of mainstream researchin Western Europe and North America, it is advisable to use Web of Science or Scopus for certain scientific disciplines (for other disciplines, no database is suitable for this purpose).

The term “preprint” is actually used for two related, but still slightly different, things. The term can refer to an author’s original manuscript (of an article, a book chapter, or a complete book) as it is submitted for publication (hence also known as “the submitted version” of the text). This submitted version typically remains private, whereas later versions of the text (revised after peer review and/or copy-edited by the publisher) are made available, either behind a paywall or in Open Access. However, the term preprint can also refer to the first public version of a text, which is being disseminated before formal peer review took place and which afterwards might or might not be developed into a more traditional publication. This second meaning of preprint is thus basically identical to what is known as “working papers” in disciplines like economics, law, and political sciences. To put it succinctly: the first meaning of the term preprint refers to a manuscript of an article, a chapter, or a book before publication; the second meaning – typically only used for articles – is considered to be the first public version of a text and therefore oftentimes treated as a publication in its own right. Both meanings of the term have in common that they refer to a text which was not submitted to formal peer review (yet).

Lately, the second meaning of the term preprint has become more dominant, not in the least because the habit of disseminating articles before they have been peer-reviewed is becoming more widespread.

Preprints can be distributed through designated preprint servers, i.e. online repositories where researchers share articles before they have undergone formal peer review. Preprint servers are often connected with a specific discipline, such as medRxiv (health sciences) or bioRxiv (biology), or region, such as AfricArXiv, and typically guarantee some basic form of quality control such as a plagiarism check before the text is accepted for publication as a preprint. However, preprints can also be shared using general repositories which are discipline-agnostic (like Zenodo) and/or platforms which accept all kinds of research outputs (such as the CORE repository of Humanities Commons), and which do not perform such basic quality checks. Preprints typically get a permanent identifier (such as a DOI) and are indexed by services such as Google Scholar, Open Science Framework (OSF) Preprints, or Web of Science’s Preprint Citation Index.

As said, the practice of disseminating preprints is on the rise. In some disciplines, such as astronomy and mathematics, up to 35% of articles start out as preprints, which are seen as an important instrument for Open Scholarship (as preprints can always be shared openly), as a way of speeding up research (since dissemination of research results is no longer slowed down by pre-publication peer review) and as a way to establish priority of discoveries. Preprints also make other innovations in scholarly communication possible (such as open peer review or the publish-review-curate approach – topics which deserve a blog post of their own) and put into question the exorbitant prices of journal subscriptions or article processing charges. Recent research by Brierley et al. and Davidson et al. in the context of the COVID-19 pandemic even brought to light that differences between preprint and final versions of articles published in biomedical journals are limited, which gives cause to reconsider the time and money spent to develop a preprint into a journal article.

Recommended reading:

J. Bosman et al. (2022), New Developments in Preprinting and Preprint Review, DOI: 10.5281/zenodo.7040997

K. Hettne et al. (2021), A Practical Guide to Preprints: Accelerating Scholarly Communication, DOI: 10.5281/zenodo.5600535

L. Mesotten – J. Berckmans (2022), To preprint or not to preprint? KU Leuven researchers share their thoughts on the (dis)advantages of preprint publishing, https://www.kuleuven.be/open-science/what-is-open-science/scholarly-publishing-and-open-access/schol-pub/interview-preprints

The common way readers use library catalogs is to find out whether they have access to a particular resource through the library they are using. Catalogs of academic libraries thus provide information about what publications a student or staff member of a particular university has access to, and how they can access it – either by providing details about the location of a physical copy or by providing a link to the electronic version.

But what is the role of the catalog of academic libraries in an Open Access world, especially now that more than half of new journal articles appear in Open Access and more and more scholalry monographs are published openly as well? If the catalog only lists what the library is paying for, then it is no longer doing its job since students and staff members actually have access to a whole lot more than what is in the catalog (namely all Open Access publications as well). If, on the other hand, the catalog lists everything that students and staff members have access too, then it becomes massive, because it should incorporate all Open Access materials as well. It also becomes rather useless since the catalog of let’s say Leuven would not be that different from the catalog of let’s say Leiden – so why would we spend any time and energy keeping separate catalogs?

Of course, if you rethink the catalog of an academic library as a curatorial instrument listing publications which subject specialists have selected as particularly relevant for a specific research community – regardless whether these are publications behind a paywall or not – then they still might make sense. Or maybe we need to think further and conclude that managing a catalog is no longer the way to fulfil the traditional, curatorial role of the library, thus acknowledging that academic libraries’ role in both discovery and fulfillment have diminished and it is high time to focus on other tasks?

The following is the redacted text of a statement given by Demmy Verbeke at the “The Diamond Open Access Model: what impact on research?” webinar organized by Academia Europaea Cardiff, KU Leuven Libraries and the Young Academy of Europe on March 28, 2022. 

Academic libraries have a responsibility in the context of Diamond Open Access on at least two levels.

For more than a decade now, librarians worldwide have played a role in promoting OA, explaining the various options to make academic work freely available to all, highlighting the pros and cons of the various routes towards OA, etc. This advocacy work is lately more and more interwoven with talking about funder compliance or talking about things like block grants, OA funds and read-and-publish deals. However, we need to be very careful that the latter does not turn librarians into salesmen for the publishers with whom their universities have this kind of agreement. The thing that we always need to remember is that academic librarians do not work for publishers; they work for their institutions and serve the scholarly community, so they need to talk about the diversity of OA possibilities. They owe it to their profession to provide an analysis which is as objective as possible of the pros and cons of various OA approaches so that authors can make up their own mind about whom they want to entrust with the dissemination of their research results.In that context, it is important that librarians also talk about Diamond OA and give the full picture. For instance by not only talking about the main thing that scholars associate with Diamond OA, namely that this is an approach to scholarly publishing which does not charge fees to either authors or readers, but also to stress the second element of the characterization used in the recent Action Plan for Diamond Open Access, namely that these are community-driven, academic-led and academic-owned publishing initiatives. This is important, because this makes an essential difference in the financial model behind initiatives of this sort and is the reason why scholars, funders and institutions alike should not only foster Diamond OA but should even prioritize it over other approaches.

The second responsibility is to not only talk and inform about Diamond OA but also to financially support it. Personally, I have very little patience for the argument “we do not have the budget to support Diamond OA programs”. Most university libraries in the Western world have multi-million budgets, whether they receive additional block grants for OA or not. I find it hard to believe that it would be impossible to find a few grand in that budget for Diamond OA. I do, however, understand and sympathize completely with the realization that we need to rethink our budgets in order to make room for Diamond OA. Both acquisition and cataloguing processes of libraries are still completely geared towards either the traditional model of publishing behind a paywall or towards publishers who have found a way to shape their OA offer in such a way that it almost appears as business as usual, for instance through read-and-publish deals. As a result, there is a big risk that library budgets are completely hoovered up by a combination of buying paywalled content and spending money on the privately-owned, for-profit approach to OA. This means that, if libraries want to financially support Diamond OA, they need to either prioritize it in the sense that they first spend available budget on Diamond OA, then on paywalled content, then on for-profit OA; or that they need to make much clearer distinctions in their budgets and need to separate a percentage for Diamond OA, a percentage for paywalled content and a percentage for for-profit OA. The added task is that they also need to stick to that division. If the price tag of either of those three categories increases – and, by the way, I guarantee that the price tag for for-profit OA will increase – then they cannot move around money from one category to the other without first having a thorough discussion that this implies a policy change.

I consider both responsibilities for academic librarians in the context of Diamond OA as an obvious continuation of the role they have been playing in the field of scholarly communication for generations. Librarians are not in the business of telling researchers what to do and how to distribute the results of their work. But that does not cancel out the fact that researchers turn to librarians for guidance in this, either by making explicit appeals to the expertise within libraries to provide support and advice, or implicitly by observing which choices libraries make in their collection building and adapting their own publishing practices to this. Similarly, research libraries have a long tradition of funding the market for academic publishing. Library budgets pay for the acquisition of monographs, for standing orders for series and for subscriptions to journals and databases. So it is natural that these same libraries are now called upon to act as funders for publishing in OA. And just like librarians were entrusted to make wise budget choices in a traditional system of acquisition of content behind a paywall, they should now be entrusted to make wise budget choices in how to support OA publishing. I, for one, am convinced that librarians will ensure much better value for money, and thus do a much better job for their institutions and the scholarly community which they serve, if they favor academic-led approaches towards OA without author fees over for-profit approaches towards OA based on publication-level payments.

A previous post shed some much needed light on the complex world of OA terminology. It certainly helps to be familiar with these names, although we cannot claim that all possible questions can be answered with one post. What about Black OA, Rogue OA, Radical OA and Platinum OA? Can Bronze OA actually be called OA? And is Elsevier really talking about the same approach to Diamond OA as the early advocates for community-driven, APC-free OA do when they plan a Diamond OA Journals conference?

It is necessary to get familiar with this OA terminology. However, we should also be mindful of the risk that we spend so much time and energy on definitions that we lose sight of the essence – not unlike what happens with trying to define what a predatory journal is, which deflects attention from the real problems of scholarly publishing (version of this article openly archived here). Another approach would be to start the discussion from two simple distinctions.

The first is the difference between open archiving and open publishing.

• The actor with open archiving is the author. He or she publishes something, and then takes the additional actions of archiving an electronic version of that publication in a repository and making the archived version freely available to all. Quite regularly, the version which the author has archived and made available is not identical to the published version (e.g. an accepted version instead of the published version), and the archived version is not available at the same time as the published version (e.g. it is made openly available with a delay of 12 months). Open archiving thus offers the advantages of OA, at no cost to the author (who can always find a free repository to archive his or her text in), but requires an additional action by the author and exists as a system parallel to actual publishing.
• The actor with open publishing is the publisher. At the time of publication, the publisher immediately makes the final version of the text openly available. A publisher typically does not do this for free, and somebody needs to cover the publication costs. This can be done by charging the author, by charging an academic institution or a research funder, or by having the publication costs covered by a group of supporters, typically a consortium of university libraries.

The second is the difference between a for-profit and a non-profit approach to scholarly publishing.

• In a for-profit approach, the goal is to realize incoming funds which are higher than the actual publication costs. The profit which is thus realized is not reinvested fully in the scholarly community but used to reward shareholders in the publishing business. Scholarly publishing has great potential for being a profit-bearing enterprise, because most of the skilled workers in the production chain (i.e. the researchers) offer the fruits of their labor (producing manuscripts, performing peer review, editorial work) for free, because there is a stable market of customers (i.e. university libraries), and because prestige is such a big factor in academic publishing (so that publishers who have attained a good reputation can realize very high mark-ups).
• The non-profit approach rejects the premise that profit should be made on the dissemination of research results. At its core is the conviction that scholarly knowledge is a common good and thus should be shared by all. It resents the fact that a small group (i.e. shareholders of a publishing company) would profit from investments with public money (both in employing researchers and by subsidizing university libraries) and therefore maintains that if incoming funds are higher than publication costs, they should be reinvested in the scholarly community.

More and more institutions and consortia of libraries conclude so-called read-and-publish deals or transformative agreements with legacy publishers. But this new incarnation of the big deal is not without its critics. The hard line opposition argues that transformative agreements hamper  progression and should therefore be avoided at all cost. A less radical approach is to make sure that the available budget is not spent exclusively on transformative agreements but is also used to support alternatives, fostering diversity of business models in the market of academic publishing.

The hard line

Transformative agreements diminish rather than stimulate diversity and equality in scholarly communication, are unnecessary in certain disciplines, might worsen the state of the market, and stimulate vendor lock-in.

Let’s look at the last argument in a bit more detail. If the negotiations leading towards a transformative agreement are successful (which is only possible if they are very well prepared – which comes at great expense), they might lead to a deal with a legacy publisher including OA at about the same cost as an earlier subscription arrangement. Hoorah! However, by concluding such transformative agreements, academic institutions demonstrate that they are able and willing to pay above production cost for publication services. What is more: by doing so, they finance legacy publishers to further develop services concerning research data management, bibliometrics, and other aspects of scholarly communication.

So what will happen next negotiation round? If (and this is again a very expensive if) negotiations go well, academic institutions might even be able to drive down the price for publication services offered. But they will have to pay additionally, and handsomely, for the other services. They will feel obliged to do so (1) because these legacy publishers will dominate the market place even more than before, (2) because the services these publishers offer will be more attractive and user-friendly than anything else on the market (since academic institutions, unwittingly but generously, provided the budget to develop them), and (3) because legacy publishers will be able to lure academic institutions into new forms of big deals packaging services concerning scholarly communication which will seem easier and cheaper than obtaining these services separately.

For this, and many more reasons, transformative agreements should actually be considered librarian malpractice.   

What if academic institutions would invest the budget as well as the time, energy, and talent they currently waste on transformative agreements in community-owned alternatives? Alternatives that foster diversity rather than monopoly and support bibliodiversity and multilingualism, thus providing a more global and democratic approach. Alternatives which involve working with partners who do not insist on vendor lock-in and who operate in service of the academic community (rather than in the service of their shareholders). Would that not mean that we would finally see, in the words of Eloy Rodrigues, the return of universities and scholars “to the driver’s seat of scholarly communications”?

Back to reality

Even if it seems a naïve dream to expect a general commitment to this approach, is it not smart to safeguard part of the budget to invest in alternatives, thus keeping the market healthy and our choices open? Even when we do it in a very small way, let’s say – as argued by David W. Lewis – by putting aside 2,5% of the total library budget to support open and community-owned infrastructure (and if you don’t know where to start, the Global Sustainability Coalition for Open Science Services can most certainly help). An investment of 2,5% seems little, and perhaps not something to be proud of (since the implication is that you spend 97,5% of your budget on scholarly communication infrastructure which is closed and/or privately-owned). But it is a start.