Over the past decade a vibrant ecosystem of so-called alternative open access publishing platforms has emerged, many of which aim to tackle some of the perceived issues with the journal publishing system other than cost. Some of these platforms represent a move away from the traditional journal as an organising principle. Journals have a number of functions within the scholarly system, from acting as certifiers and a public record of who did what and where; acting as gatekeepers to that record via editorial selection and peer review; and disseminating work to different audiences. Alternative platforms sometimes seek to disaggregate these functions. They might also differ from traditional scholarly journals in other ways, such as their publication process, governance and underlying infrastructure. They often apply a wider disciplinary scope, include the publication of submitted versions/preprints or outputs other than traditional ‘articles’, ‘monographs’ or ‘books’, and apply open and/or post-publication peer review. Often the focus is on free availability of content, transparency and efficiency, or on changes to improve the intrinsic quality of the research work (such as the format of Registered Reports, where the methods are peer reviewed before any experimental work is undertaken), rather than selectivity, impact or prestige.
In 2022, the Knowledge Exchange1 started a project, named ‘Alternative Publishing Platforms’, the details of which are set out in a scoping paper.2 The aim of the project is to gain a better understanding of the landscape formed by these platforms and how they can be placed in the open scholarly communication ecosystem. It also aims to provide research stakeholders (researchers, funders, research performing organisations, libraries) with information to help them identify opportunities for political and financial support for strengthening the sustainability of these platforms.
2. Alternative Publishing Platforms
The scoping paper, circulated in April 2022, proposed defining Alternative Publishing Platforms on the basis that “they often apply a wider disciplinary scope, include the publication of submitted versions/preprints and apply open peer review” and that “often the focus is on free availability of content, transparency and efficiency rather than selectivity or prestige.” These platforms could also focus on speed of publication, on reforming the peer review process, on reproducibility and replicability of research results and on publication bias.
The examples given in the scoping paper enabled us to identify an initial typology consisting of 1) funder platforms, 2) stakeholder governed platforms and 3) experimental publishing platforms.
The use of the term "alternative" was recognised as presenting the risk of a certain narrowing, or even ambiguity, particularly in the case of platforms that are mainstream in their linguistic or disciplinary area (e.g. Scielo). However, this term, despite its shortcomings, and as acknowledged in the scoping paper, allows easier designation of the object under study in order to make contrasts to the legacy stakeholders of academic publishing.
3. Survey
Further work, based on the scoping paper, consisted of drawing up a survey aimed at Alternative Publishing Platforms, to identify and better understand the individual platforms. This led to a call for entries, and which is still open, to any platform working in open access publishing / scholarly communication who wishes to be included in the Knowledge Exchange taxonomy of Alternative Publishing Platforms.
This survey was designed over the summer of 2022, with the aim of asking the platforms about all the areas and functionalities related to their activity. With the data thus collected, a landscape analysis has been conducted, together with the development of a prototype visualisation enabling to manipulate and showcase the data for end-users.
A first version of the questionnaire was tested in November 2022 with 13 platforms based in Knowledge Exchange countries (Germany, the Netherlands, France, Denmark and the United Kingdom) as well as Sweden, with which the members of the working group were already in contact. This test led to a number of adjustments, resulting in a 28-question questionnaire (appendix Questionnaire APP). The call for entries was launched publicly on the 14th of February 2023 and broadly disseminated.3 Over the following two months, 45 responses from different platforms were received.
The survey continues to stay open and new responses are welcomed, but for the purposes of this report, work has focused on the 45 responses received up until the 4th of April 2023.
4. Methodology
The questionnaire was disseminated widely, including via social media, as well as being sent specifically to contacts identified by the working group. This means that, although a number of platforms were identified by the members of the working group as falling under the definition of ‘alternative publishing platforms’ and invited to take part, particularly for the questionnaire test phase, the corpus analysed in this report includes all the responses received, including those who self-identified as falling under this definition.
The data collected as part of the survey has been processed for analysis:
Firstly, consolidation work was performed including de-duplication and harmonisation, particularly for platforms that responded to both the test phase and the final questionnaire.
Secondly, an analysis of the responses for which free text was requested, as well as the 'other' responses for which details were provided. This led to some responses being completed on the basis of the free text analysis.
We have also established a classification to distinguish three main types of platforms: those aimed directly at authors and readers (author/reader facing), those hosting academic journals (author/reader facing & tech stack) and those offering technological solutions (tech stack). For more details on this typology, we refer to the presentation of the data visualisation prototype (chapter 6).
Lastly, the nine thematic blocks identified and presented below in the "Results" section were analysed and cross-referenced with other criteria (e.g. the number of responses, the number of questions, etc.).
The consolidated data of the 45 responses received is available in open access on Zenodo: 10.5281/zenodo.8358919
5. Results
5.1 Disciplines and type of publications
Since the 18th century, research communities have developed greater and greater degrees of specialisation, accompanied by increasingly specialised publication outlets and outputs (for a review see, for example, Stichweh, 20014). Changes to the trajectories of these publication systems, such as the move to open access, has happened differently in different disciplines (for example, see Severin et al., 20205). In this survey we hoped to identify whether there were differences in the way that different fields had embraced ‘alternative’ publishing practices, or indeed whether alternative platforms often sought to be more discipline-agnostic (possibly as a way to break down silos).
The platforms were asked which research disciplines they were designed for. 45 platforms answered the question and a majority of platforms said they were designed to cover all disciplines: 29 platforms (or 64% of all answers) answered ‘All’ to this question. Of the 29 who said that they served ‘all’ disciplines (at least in principle), 7 were author-facing: ScienceOpen, ResearchEquals, PubPub, Peer Community In, F1000 Research, Humanities Commons, Octopus. The rest were tech stacks or journal hosting platforms. All six of those who said they served both social sciences and humanities were journal hosting platforms.
A smaller group of platforms stated that they only cover one or more of the specific disciplines. Four platforms said that they cover ‘Physical Sciences/Technology’, 5 platforms cover ‘Life Sciences/Medicines’ and respectively 9 and 7 platforms cover ‘Social Sciences’ and ‘Humanities’ (multiple selections were allowed). See Figures 1 & 2.
Fig 1. Research disciplines which each platform in the survey said that they served (n=45).
Fig 2. Research disciplines which each platform in the survey said that they served (n=45).
We also wanted to investigate how many platforms cover publication types that are less often associated with traditional journal publishing e.g. lab notebooks and software & code.
Most platforms publish research articles (38 of 45 platforms). Next in line are proceedings - 26 platforms publish proceedings. This is followed by books and edited collections that are published by 23 of the platforms. See Figure 3.
BooksFlow, CBS Open Journals, Event Notifications in Value-Adding Networks, Dynamica by SciFree, Episciences, F1000 Research, Humanities Commons, NumeRev, OpenEdition, PeerCommunity Journal, PressBooks, PsychArchives, PUBLISSO Platform Gold, PUBLISSO System, PubPub, ResearchEquals, SciELO, ScienceOpen, SciPost, and tidsskrift.dk all said that they published some of what might be termed non-traditional publication formats, such as datasets, blogs, software etc. as well as traditional research articles or proceedings. However, many of these are tech stacks (i.e. they have the potential to publish ‘nontraditional outputs’), rather than author-facing.
Of the 7 platforms that don’t publish research articles two platforms said that they publish pre-prints (Open Preprint Systems and Peer Community In, although Peer Community In publish recommendations and reports based on preprints rather than the pre-prints themselves). These platforms, therefore, are essentially still built around traditional research articles, although they are innovating with aspects of peer review and dissemination. There are 2 platforms who said that they publish books and edited collections (AU Library Scholarly Publishing Services and Open Monograph Press). Only two offer solely ’non-traditional’ formats (such as datasets, protocols, proposals): Octopus and Sciflection.
Most of the alternative platforms, therefore, are still ‘alternative’, but perhaps not to the extent or in the way initially imagined, since they are still largely rooted in traditional article publishing, even though many are offering opportunities to publish non-traditional materials as well. The platforms that are author-facing and offer non-traditional outputs are: Octopus, SciPost, PsychArchives, F1000 Research, Humanities Commons, ScienceOpen, ResearchEquals, PeerCommunity Journal & PubPub.
Fig 3. What publication types are covered (not necessarily hosted) by the platform? (n=45)
5.2 Publication functions
In 1664, right at the birth of the scientific journal, Henry Oldenburg, secretary to the Royal Society in London, proposed a publication (later called the Philosophical Transactions) in a letter to Robert Boyle on 24th November6:
“The Society always intended, and, I think, hath practised hitherto, what you recommend concerning the regist’ring of the time, when any observation or expt is first mentioned, and they upon this occasion have declared it again, if it should be punctually observed: in regard of work Monsieur de Zulichem [Christiaan Huygens] hath been written to, to communicate freely to the Society, what new discoveries he maketh, or what new expts he tryeth, the Society being very careful of regist’ring as well the person and time of any new matter, imparted to you, as the matter itself; whereby the honor of the inventor will be inviolably preserved to all posterity.”
Between them, Oldenburg and Boyle in these letters set out what were to become seen as key functions of journals: registration (‘whereby the honour of the inventor may be inviolably preserved’), certification (establishing the validity of work), dissemination (sharing it with others) and archiving (whereby it can be ‘preserved to all posterity’).7
45 platforms answered this question and almost all platforms cover dissemination of publications (43 platforms), while nearly as many platforms cover archiving (39 platforms). See Figures 4 & 5.
When it comes to registration and certification the numbers are a bit different. A little more than half of the platforms say they cover registration (26) and 23 platforms say they cover certification.
Fig 4. Which publishing functions are covered by the platform? (n=45)
Fig 5. Which publishing functions are covered by the platform? (n=45)
While almost all platforms owned by academic institutions offer dissemination and archiving, only about half of them offer registration and certification. Almost all not-for-profit platforms offer all four functions, while only half of the for-profit platforms offer registration and certification.
5.3 Ownership and governance
When we initially defined what characterises alternative publishing platforms, one aspect was that it could be seen as an alternative to the traditional commercial publishers. It is therefore interesting to look at who owns such platforms.
The majority of the responding platforms said that they are owned by academic institutions (24), and only 8 platforms are owned by for-profit companies: Booksflow (by Progettinrete), Dynamica by SciFree, F1000Research (by Taylor and Francis), OPUS, Pressbooks (by BookOven, Inc), ResearchEquals (by Liberate Science GmbH), ScienceOpen, and Sciflection. 10 platforms declare themselves as owned by not-for-profit organisations. See Figure 6. If academic institutions are also considered non-profit, then 35 of 45 platforms declare themselves owned by institutions and organisations that are non-profit.
Fig 6. What kind of ownership structure does the platform have? (n=45)
The eight platforms that characterise themselves as for-profit are not limited to specific disciplines. Seven of them cover physical sciences, life sciences/medicine, social sciences and humanities.
We asked separately about the governance structure of each: is it community based, academic-led, private or institutional?
More than half (25) of the 45 platforms said that they are academic-led. Almost as many (23) platforms say the governance structure is institutional. Only 10 described themselves as privately governed. See Figure 7.
Fig 7. What is the governance structure of the platform? (n=45)
5.4 Publication chronology
Some publishing outlets offer "continuous article publication": releasing articles (or other materials) individually as they are accepted, instead of waiting for all articles intended for a specific issue to be ready. This practice enables faster availability of (peer reviewed) research results. In addition, there is a shift from the version of record (VoR) towards a record of versions8.
Other outlets allow ‘living articles’ and reversioning (e.g. ’living systematic reviews’9), whereby the information in an article can be updated post-initial-publication. As both approaches (continuous publication and versioning) seem to be adopted more and more, we included the question “Does your platform offer continuous publication (subsequent versions published on the platform or publication continually updated)” into our survey. We received 44 responses to this question. 28 platforms stated that they are supporting these practices, while 17 platforms stated that they do not offer continuous publication or versioning. Seven platforms are planning to introduce one or both of these features as they answered “not yet”.
In the next question, we asked the platforms to further specify their use of continuous publication and/or versioning. 24 platforms provided respective free text answers. Most answers referred to “versioning”: 17 answers explicitly mentioned new versions/updates of publications or data and named e.g. concepts as living handbooks. The ability to update research outputs seems to be an important feature for many alternative publishing platforms. In contrast, the topic of continuous publication was only addressed in three of the free text answers explicitly, perhaps highlighting the move away from the concept of ‘issues’ which is so tied to the historical process of producing a print edition of a journal. Three answers highlighted other aspects in conjunction with their publication model.
Comparing these results to Q18 (where we asked whether the content selection for a platform is based on “quality of the methodology” or the “expected impact of the research results”), more platforms that select their content based on the quality of the methodology (10) or that leave content selection to editors (14) also offer continuous publication than those explicitly selecting content based on the expected impact of the research results (4).
With our question on whether submission is initiated by the “submitting author” or a “requesting editor” we aimed to approach the topic of overlay journals. Overlay journals assemble their content from already openly available sources online (such as preprints). This question was widely misunderstood: some respondents understood that the question referred to the “power to press the publish button” (e.g. preprint publication without any mediator) while others interpreted it in the “overlay journal spirit”. As respondents interpreted the question in different ways, answers cannot be evaluated//interpreted in a meaningful way.
The topic of immediate publication gains special attention in times where fast communication of new scholarly knowledge can be a game changer. During the Covid-19 pandemic, for example, preprint servers received high attention and played a crucial role in the dissemination of new findings10. This model is also a departure from the traditional pipeline which has always involved an editor and/or peer reviewer being involved in deciding on whether and when a publication is made available to readers. Hence, we asked at which point in time the alternative platforms surveyed make the research outputs public.
We received 45 responses to this question (see Figures 8 & 9). 10 platforms answered “immediately (author in charge of publication time)”, 18 respondents chose “After initial checks (editor/platform in charge of publication time)”, and 36 answered “After peer review (peer reviewers/platform in charge of publication time)”. Both models - “immediate publication” (with or without checks) and “publication after peer review” - are almost equally present. Furthermore, adoption of the models is not an “either/or” question but implementation seems to be often in parallel or as subsequent steps.
Fig 8. At what moment does publication occur on the platform? (n=45).
Fig 9. At what moment does publication occur on the platform? (n=45).
When breaking down these findings by academic disciplines, it is first to be noted that platforms often are publishing in more than one discipline. The majority of responses from the humanities and social sciences state that they publish “after peer review”, but a clear trend for immediate publication cannot be observed for any discipline. Furthermore, the majority of research articles published on the surveyed alternative publishing platforms undergo peer review before publication. This also holds true for proceedings publications.
Out of the 23 platforms offering something more than the traditional ‘after peer review only’ model, only 7 are author-facing: ScienceOpen, ResearchEquals, PsyArchives, Octopus, Humanities Commons, F1000 Research, and Beilstein Journals/Archives.
5.5 Peer-reviewing
Peer review is an aspect of scholarly publishing that has evolved a great deal during the past few centuries.11 The original journals tended to rely on an editor to decide what to publish, or an internal committee, in many cases part of a scholarly society. Einstein was famously horrified to find that a journal editor had sent one of his papers to an external referee and withdrew it - none of his other papers were refereed.12 In the 1930s, papers in the journal Science were still mostly solicited personally by its editor, and not peer-reviewed - although one paper he deemed outside of his expertise he asked his son to read (who rejected it).13 Many of the current journals adopted external peer review in the mid-20th century, and it really only became systematic in the 1990s as the volume of papers submitted became difficult to deal with.
This system is showing signs of dysfunction in the 21st century, and different models, such as post-publication and open peer review are now used in some systems. Post-publication peer review can help work get published faster, but removes any form of quality gate-keeping that pre-publication peer review was designed to give.
We asked platforms whether peer review in their system was pre-publication, post-publication, at multiple stages, editor choice only, or not applicable. See Figure 10 & 11.
29 selected pre-publication; 13 post-publication only; 9 multi-stage; 2 mentioned this was the editors’ choice and 2 indicated it was not applicable to their platform. This shows that a small minority are using the pre-1970 model of editor choice, about half of these platforms are using the late 20th century pre-publication peer review model, and about half are doing something different: either purely post-publication or peer review at a variety of stages.
Fig 10. Which peer review stages are offered at the platform? (n=45)
Fig 11. Venn diagram of answers to the question ‘which peer review stages are offered at the platform?’ (n=45). For the Venn diagram, the category ‘multi-stage peer review’ was considered to be both pre- and post-publication peer review, and has been added to the diagram accordingly.
Those selecting purely post-publication were: Octopus, CBS Open Journals, PUBLISSO Journal and Platform Gold, Event Notifications in Value-Adding Networks, ResearchEquals, PsychArchives, F1000Research, Humanities Commons and Opus Journal.
Whether the peer reviews are done double-blind (neither reviewer nor authors know each others’ identities and the reviews are kept confidential), whether one party knows the identity of the other, or both are transparent also very much affects the culture around reviewing.
Purely anonymous peer reviewing aims to allow frank and unbiased comments. Purely transparent peer reviewing can have multiple aims, including ensuring accountability for comments and allowing other readers to benefit from the reviewers’ comments.
Historically, peer reviewers have tended to be anonymous, and the paper’s original authors are sometimes also anonymous at the point of review.
In our survey, 15 of the platforms at least sometimes had fully transparent identities (only 8 were exclusively fully open), and 16 had full anonymisation at least sometimes. 20 reported that peer reviewers could be anonymous but authors’ identities not hidden, 8 that authors could be anonymised but peer reviewers not. 8 reported that it was up to editors what kind of anonymity was maintained (see Figure 12).
Fig 12. What kind of transparency of peer review does the platform offer? (n=45) Multiple answer selections allowed
This spread shows interesting diversity highlighting the exploration of different peer review systems.
Those platforms offering purely open reviewing are: ResearchEquals, AU Library Scholarly Publishing Services, peer community journal, Dynamica (by SciFree), F1000Research, Sciflection, SCIndeks: Serbian Citation Index, and Octopus. These are all domain-agnostic platforms.
Finally, we asked platforms whether or not they published their peer reviews (see Figure 13). Traditionally they have not been published, and 24 platforms reported that they did not publish them - with an additional 2 saying that they would publish them if the reviewers agreed and 6 if the authors agreed. 16 reported that the reviews were always published (8 of them only if the article was published, in systems that had a gateway model). 10 reported that it was up to the editors if a peer review was published, and 21 reported ‘other’ systems. Some mentioned that they were actively considering other forms of peer review publication in the future. This again emphasises the diversity of peer review models being explored.
Fig 13. What kind of availability of peer review(s) does the platform offer? (n=45) Multiple answer selections allowed
5.6 Content selection
Peer review was developed as a way of helping filter publications on some measure of perceived quality when interest in publications outstripped the available supply of space in journals. The format of scholarly publications and measures of quality have changed considerably over the past few centuries, and so we were interested to know what criteria of quality were perceived as important by different platforms.
The survey options for selection criteria were perceived impact of the findings (the ‘high impact publication’ of the later 20th and early 21st century), the perceived quality of the methodology (a focus of some early 21st century publication systems), ‘up to the editors’ or ‘other’.
Seven reported that they focussed on expected impact (2 exclusively so) plus 1 (exclusively) on Altmetric scores (OOIR), a form of impact measurement (reported under ‘other’). 16 reported focussing on methodology (9 exclusively so), 17 that it was ‘up to the editors’ and 6 ‘other’. Some respondents mentioned subjective interest of the editor, addition of knowledge to the field or diversity of the platform. In addition, some platforms had no barriers to publication, and thus did not employ selection. See Figures 14 & 15.
Fig 14. On what basis are publications on the platform selected or rewarded ? (n=45)
Fig 15. On what basis are publications on the platform selected or rewarded ? (n=45)
Those focussing solely on the quality of the methodology were: PUBLISSO Platform Gold and System, Centre Mersenne, tidsskrift.dk, AU Library Scholarly Publishing Services, F1000Research, OPUS, Beilstein Journals/Archives, and Octopus.
5.7 Revenue models
The question of the platform's business or revenue model is linked to the issue of sustainability. This is all the more important for platforms that want to move away from for-profit models. Since 2020, the issue of the sustainability of open access academic journals without APCs (the diamond model) has given rise to several initiatives in Europe and internationally (OA Diamond Study in 202114; Diamond Action Plan in 202215; DIAMAS project in 202216). Alternative publication platforms face the same challenges. A first step is to analyse the sources of revenue they use to support their operations.
17 platforms said they have only one source of funding, and the maximum number of sources indicated is 6 (1 platform). The table below presents the distribution of the platforms against the number of revenue types. 5 platforms indicated that this question was not relevant for them because it was up to journal editors.
Table 1: Presenting the distribution of the platforms against the number of revenue types.
Fig 16. What is the financial and/or sustainability model for the platform ? (n=45) Multiple selection allowed.
The most common source of funding is the host institution or platform owner covering all or part of the costs (25 out of 45). Funding by grants and by a consortium of stakeholders (11) are the next more frequent sources, followed by government funding (9). See Figure 16.
Three sources of funding are clearly in the minority:
Payment by the reader is mentioned by two platforms, both in HSS. This source of revenue is always offered in addition to other streams. In the case of OpenEdition, this is the implementation of the Freemium model, where the HTML version of the texts is open access, while access to certain formats of the same texts requires a subscription from the libraries. As far as Booksflow is concerned, no details have been provided, but it is interesting to note that this is a Tech Stack platform.
Payment by authors is mentioned as a source of income for 6 of the respondents. With the exception of ScieLO, which is a Journal Hosting platform, all are Tech Stack platforms whose funding also relies on other funding streams as well.
Finally, 6 platforms rely on the financial participation of a partner publisher. These include four Journal Hosting platforms (Episciences, Mersenne, Open U Journals TIB Open Publishing) which require a financial contribution from the journals they host, two Tech Stack platforms and one platform (SCIndeks) for which this is the only source of revenue indicated.
Respondents were able to indicate other sources of funding. One mentioned advertising (Sciflection) and another voluntary donation based on the choice of the licence (ResearchEquals, see above).
The heatmap below (Figure 17) shows that the revenue streams most often combined are “Host institution / platform owner pays” together with “Grant to platform pays” (9 combinations) followed by the pairs “Host institution / platform owner pays” + “Stakeholder consortium pays” and “Grant to platform pays” + “Stakeholder consortium pays” with 5 combinations for each.
Fig 17. What is the financial and/or sustainability model for the platform - combination of different revenue types (n=45). Multiple select allowed.
The 8 for-profit companies that responded all have a mix of revenue streams (including ‘Author pays’ for 4 of these 8, against 6 of the total 45) except for ResearchEquals. They are also all tech stack platforms except Research Equals.
5.8 Technology and interoperability
Publishing software and technology has evolved tremendously over the past years and there are now many variations and service providers that offer online publishing software, modules and technical solutions. Whereas in the early days of digitisation in the mid-1990s it was mainly existing publishers that invested in online publishing platforms, these have now mostly been overtaken by newer commercial publishers and innovations in Open Scholarly Communication and initiatives originating from the academic community itself. For example, Open Journal Systems developed by the Public Knowledge Project (PKP) was already available in 1998, and is now in version 3. Its code and software is fully open source, and used by tens of thousands of academic journals and journal platforms worldwide.
‘Open source’ in this particular survey stands for open licenced source code of the publishing software that is made freely available for reuse and modification through forking and redistribution via a software repository. The opposite would be proprietary (commercially developed) software. This could mean the only user is the publishing platform itself, though there are also plenty of examples of publishing software that have been white-labelled and offered to third parties for a fee, like for example Sciendo from De Gruyter.
In the survey we only made the distinction between either open source software or proprietary software.
Fig 18. What type of software is the infrastructure based on? (n=45). Multiple select allowed.
33 respondents stated that they run or make use of open source software. 28 of those 35 respondents also submitted the information where the code and software is stored and made available. 14 of those respondents made use of PKP software, either original software versions or a derivative. A correlation can be seen of those respondents who are open source and the type of ownership (comparing Q6 with Q21). The vast majority that offer their publishing software open source are embedded in or managed by an institutional entity (22 out of 33) or are otherwise organised as non-profit organisations (8 out of 33). One privately owned publishing platform (ResearchEquals) has made its code openly available. The majority of those responses can be connected (comparing Q20 with Q21) to the way these publishing platforms operate and are being financed. In most cases this is the hosting institution (17) or a stakeholder consortium which sustains the platform (10). See Figure 18.
Some answers show an overlap, meaning that respondents ticked both boxes. They were mostly in principle open source, but depending on proprietary software solutions, such as specific hosting services (e.g. PubPub).
The same correlation can be seen with the question about whether the publishing platforms are offering an open API and/or enhancing interoperability with other services (Q23) - see Figure 19. An API will make basic information on websites and in webtools easier to reuse and integrate in other services of external parties, e.g. universities' registration systems, ORCID, and other (research) tools. The majority that offer an open API are owned by an institutional entity (14 out of 28 responses) or are otherwise organised as non-profit organisations (6 out of 26). We also asked whether the platform actively works on enhancing interoperability with other services. A majority is actively offering or working on interoperability with other services. Examples that are mentioned most are Crossref (15), DOAJ (10) and OAI-PMH protocol (7).
Fig 19. Does your infrastructure have the following ? (n=45). Multiple select available.
5.9 Copyright and licensing
The licence applied to an academic text or research data, code and software plays an important role in the ability of users to appropriate, redistribute, enhance or adapt it. In this regard, the launch of Creative Commons licences in 2002 represented a major turning point by adapting the legal framework to the open dissemination of knowledge on the web17. These licences allow permission to be granted in advance to use a work in contrast to the traditional “All rights reserved” mode of distribution. The world of higher education and research quickly saw the potential of this new type of licence which has continued to develop over the last 20 years.
We therefore looked at the licences offered to authors by the platforms, making a distinction between open licences (Creative Commons and GNU) and cases where all rights are reserved. Of the 45 responses to this question, there were 135 possibilities for disseminating content under open licences and 3 possibilities for reserved dissemination, indicating a strong preference on the part of these platforms for licences that favour an open approach to the circulation and re-use of scientific knowledge. 9 of the respondents stated that the choice of licence was a matter for the journals hosted and not for the tool itself. See Figure 20.
Fig 20. What does your platform offer regarding licensing? (n=45). Multiple select allowed.
The three platforms offering the possibility of dissemination without an open licence, "All rights reserved", also offer other licence options, including CC BY. It is interesting to note that these three platforms (Booksflow, Pressbooks and Research Equals) are among the 8 for-profit companies that responded to the survey. Among these, the model proposed by Research Equals is particularly noteworthy: the author's choice of a closed licence is necessarily subject to a fee, whereas the choice of an open licence leaves open the possibility of not paying any fees.
Within the open licences, the CC BY licence is offered by all the platforms hosting scientific content (36). Several platforms offer the choice of using the other CC licences, the numbers ranging from 22 for CC BY-SA to 12 for CC NC-SA. Two respondents specified that the CC0 licence was applied to the metadata while the choice of the text licence was left to the choice of the authors.
There is no marked disciplinary difference in the use of the CC BY licence, which is often considered to be the most open. 100% of Physics and Life science platforms offer it, and 90% of HSS and/or multidisciplinary platforms. Only three platforms do not offer the CC BY licence option: Melusina Press (CC BY-SA), Roma Tre Press (CC BY-NC-ND) and OOIR, which has opted for the CC0 licence, which is even more open than the CC BY.
It is also noteworthy that 100% of the 12 Author Facing platforms (as opposed to Journal Hosting and Tech Stack) offer the CC BY licence option. Finally, as far as software is concerned, the GNU licence is offered by three platforms. And in the comments the Apache 2.0 licence was mentioned.
The issue of copyright management is the counterpart to that of licensing. While there is a fairly wide variety of licences offered and used, it is striking that none of the platforms that answered this question said that authors should assign exclusive copyright over their works.
For 35 platforms, the rights assigned are non-exclusive, which allows authors to implement strategies for disseminating or re-using the scientific content produced. 8 respondents, mainly technical solution providers, replied that copyright management was the responsibility of the hosted journals and not the policy of the software providers. See Figure 21.
Fig 21. What does the platform offer regarding copyright? (n=45). Multiple select allowed.
It is apparent that ‘alternative publishing platforms’ span a wide range of platforms across multiple dimensions. At surface level, they can be distinguished by their primary functionality: journal hosting platforms, author-facing platforms, tech stack and ‘others’, with the two indexing services in our survey being an example of the latter. Even in this basic classification, there are already overlaps, e.g. author facing platforms that also function as tech stack for other publishers (e.g. PubPub, ScienceOpen and F1000Research), and journal hosting platforms that in turn function as tech stack for other publishers - Publisso being an example of a platform providing both types of services.
More fundamentally, platforms in each of these categories also differ in what kind of alternative they provide to mainstream traditional publishing - this can vary from serving multiple disciplines and publication types, various forms of open peer review, timing of publication in relation to peer review, exclusive use of open licences, etc, to technical and organisation aspects such as the use of open source code, academic ownership and governance, and the business model on which the platform operates.
To illustrate how alternative publishing platforms can and do differ across these characteristics, we provide an interactive visualization of the 45 platforms that have responded to our survey, enabling users to select a range of characteristics and see which platforms use or facilitate those aspects of publishing. See Figure 22 for screenshots of this.
Fig 22 A/B Screenshots of interactive visualisation of different characteristics of the platforms included in the survey A - all platforms shown B - selection of platforms based on chosen criteria
The aim of this visualisation is to provide another view on the multiple aspects of alternative publishing platforms. It could be used to find examples of different types of innovation in scholarly publishing, and explore the variety in characteristics of current platforms. For funders and institutions considering financial and/or in kind support for alternative publishing platforms, it might be able to support a discussion on the type of publishing initiatives the organisation would find most valuable to support. This could either be based on basic classification (journal hosting platforms, direct author-facing venues or tech stack supporting multiple publishing initiatives), or on particular characteristics, like disaggregation of publishing functions, role of peer review and/or business and governance models.
It is explicitly not meant as a comprehensive overview, and/or selection tool of alternative publishing platforms, as it is only based on the relatively small sample of survey respondents in this project. In addition, the information the visualisation is based on is derived from survey responses and our interpretation thereof, and might therefore contain discrepancies. We are open to feedback on the results, as well as on entries from additional alternative publication platforms to be added to a potential future version of the visualisation.
7. Discussion
Who is innovating in scholarly publishing? Most of the alternative platforms in this survey were institution-based and driven by academic or similar communities. This could be down to selection bias as a certain number of platforms for the survey were invited by the group, and institutional platforms may be better known to the group. We emphasise that the survey is still open to data collection and would encourage any other platforms that would describe themselves as ‘alternative’ to submit data, in order to inform future research and the visualisation tool. There appears to be less innovation from within the established commercial scholarly publishing system - but again this could be due to selection bias within this particular sample.
From our sample, no pattern emerged of any discipline appearing to be more innovative than any others, and indeed most alternative platforms seemed to be open to use by all fields.
What innovation do we see in terms of publication format? Actually, most platforms within this survey were replacing the function of existing publishers in publishing research articles, books and conference proceedings, although many were attempting to broaden what was acceptable for publication (e.g. accepting datasets or peer reviews alongside traditional formats).
There was some innovation around peer review, with some platforms set up solely to host or organise peer review (e.g. Peer Community In). Several platforms exclusively used the model of post-publication peer review, and several only used open peer reviews. These may not be ‘new innovations’ in peer review, but they might certainly be considered ‘alternative’ to the established pre-publication peer review model of the 20th and early 21st century.
Considering both of these aspects, only a small group of fewer than 10 of the 45 platforms should probably be described as truly exploring ‘alternative ways of doing things’.
This might be another small bias of the study as a whole, as the truly alternative publishing platforms may have found it more difficult to fit their self-description into a survey designed by those most familiar with traditional publishing workflows.
Are we seeing platforms pushing to change the incentive structure in favour of being more selective for quality over potential ‘impact’ of the findings? Only 11 of the platforms said that they solely concentrated on the methodological quality of the work, 2 solely on the impact of the work. Most said it was up to the editors to decide on criteria for assessment - the platforms themselves were agnostic. This is an area where further work might help elucidate the philosophies of different platforms when it comes to research assessment.
Platforms in this survey which are author-facing, and which offer non-traditional outputs as well as purely open peer reviewing are: Octopus, F1000 Research, ResearchEquals, and PeerCommunity Journal. These might be seen as the cluster of ‘truly alternative’ platforms in this survey. Who funds these? Octopus and PeerCommunity Journal are both not-for-profit organisations - Octopus funded by a grant from UKRI and PeerCommunity Journal by funding from a consortium of institutions. ResearchEquals was developed with a grant, and now has a particularly innovative funding structure of ‘pay to close’ where authors can publish for free open access, but pay to publish non-open access. F1000 Research was bought by journal publishers Taylor & Francis and charges authors to publish.
8. Concluding remarks
Although there are many outlets which identify themselves as ‘alternative’ publishing platforms, the majority of those that took part in this survey retained many of the features of traditional publishers (such as output formats, pre-publication and closed peer review). The aspect that made them ‘alternative’ was more commonly their not-for-profit origins - most were founded by institutions or academic communities. Their ‘alternative’ focus appeared mainly to be in providing open access and open source code. This suggests that the main driver for alternative platforms has probably been an academic/institutional response to paywalls and perceived commercialisation of scholarly publishing, and the subsequent research policy decisions of universities, funders and for example European science and research policies on open science.
Innovations around different methods of peer review and research outputs/publishing formats are the preserve of outliers (or vanguards) of the ‘alternative’ scene, pushing boundaries further. These are often more recent innovations and it remains to be seen whether they will eventually change the publishing landscape further. Whether they do or not is probably dependent on decisions taken in research policy in the coming years.
Acknowledgments: this study was underpinned by the work of the other Knowledge Exchange Task and Finish Group members on Alternative Publishing Platforms:
Anna Mette Morthorst (DeiC), Knowledge Exchange lead
Arianna Becerril-Garcia (AmeliCA/Redylac)
Bas Cordewener (Knowledge Exchange)
Christina Hemme (DFG), Knowledge Exchange co-lead
Daniel Beucke (University of Göttingen)
Georgia Hemings (Knowledge Exchange)
Janne Tuomas-Seppänen (University of Jyväskylä)
Karin van Grieken (Knowledge Exchange, SURF), former KE co-lead
Mafalda Marques (Jisc)
Minna Ala-Mantila (CSC)
Priscilla Dibble (Knowledge Exchange)
Rasmus Rindom Riise (Copenhagen University Library / Royal Danish Library)
Call for entries: The Knowledge Exchange taxonomy of Alternative Publishing Platforms
Alternative publishing platforms represent a move away from the traditional journal as an organising principle and might differ from traditional scholarly journals in a number of ways, including publication process, governance, and underlying infrastructure. They can be regarded as examples of real innovative, open access scholarly communication or as effective "threat infrastructures" to traditional journals. These different platforms all have different aims, such as seeking to remove the barriers, constraints and costs imposed by legacy academic publishing companies, to reduce questionable practices, or make research work more deeply accessible and reusable. They also feature different ways of operation, different innovations, different business models and different structures of governance. Describing this diversity and knowing the directions in which these platforms are driving innovation, will allow us insight into what can be a confusing landscape.
In 2022, Knowledge Exchange started a project exploring what Alternative Publishing Platforms do and how they can be placed in the open scholarly communication ecosystem. In order to help guide conversations, we first published a Knowledge Exchange (KE) scoping paper on this. The next step is to identify and better understand the individual platforms. This call for entries is open to any platform working in open access publishing / communication who wishes to be included in The Knowledge Exchange taxonomy of Alternative Publishing Platforms.
The application form below is not personal – it should be filled on behalf of the entity operating the platform. Information provided in the application will be used to decide inclusion in the taxonomy, and for research purposes. Personal information (name, role, email address, etc.) about you will be treated as confidential and will only be used to verify you are entitled to give answers on behalf of the platform, and to contact you to inform you about the progress of the project and announcing project results.
Your participation is voluntary. You are entitled to ask that part, or all, of the answers you gave for the application be deleted.
* Mandatory question
1. Name of the publishing platform*
<Your answer>
2. URL of the publishing platform*
<Your answer>
3. E-mail address of the respondent (this won't be shared publicly):*
<Your answer>
4. Which research discipline(s) has the platform been designed for?*
Physical Sciences/Technology
Life Sciences/Medicine
Social Sciences
Humanities
All
Other:
5. What is the governance structure of your platform?*
Community
Academic-led
Private
Institutional
Other:
6. What kind of ownership structure does your platform have?*
Charity/association
Academic institution
Not-for-profit organisation
For-profit company
Other:
7. Under what type of legal entity does your platform operate (e.g.charitable organisation, (private) foundation, corporation, etc. - click here for an overview)
<Your answer>
8. What is the (intended) community function of the platform (readers, authors and/or editors form a community that the platform helps grow/sustain /cultivate)*
<Your answer>
9. Which publishing functions are covered by the platform?*
Registration (for claiming precedence)
Certification (for establishing validity)
Dissemination
Archiving (to preserve the record)
10. What publication types are covered (not necessarily hosted) by your platform?*
Research articles
Books and edited collections
Proceedings
Reports
Preprints
Datasets
Software & code
(Software) documentation
Patent
Research Proposal
Thesis
Poster/Presentation
Lab Notebooks
Peer reviews/recommendations
Protocols
Other:
11. Does your platform offer continuous publication (subsequent versions published on the platform or publication continually updated).
Yes
No
Not yet
12. If you selected 'yes' at question 11, please specify:
<Your answer>
13. Is publication initiated by:
Submitting author
Requesting editor
14. At what moment does publication occur on the platform?*
Immediately (author in charge of publication time)
After initial checks (editor/platform in charge of publication time)
After peer review (peer reviewers/platform in charge of publication time)
15. Which peer review stages are offered at your platform?*
Not reviewed by anyone before making it public
Peer review pre-publication
Peer review post-publication
Peer review multi-stage
Other:
16. What kind of transparency of peer review does your platform offer?
All identities shown to all participants
All identities published
Reviewer anonymized
Authors anonymized
Authors and reviewers anonymized
Authors, reviewers and editors anonymized
Other:
17. What kind of availability of peer review(s) does your platform offer?
Reviews published regardless of decision
Reviews published if article accepted
Reviews published if author consents
Reviews not published
Other:
18. Are publications on the platform selected or rewarded on the basis of:*
Quality of the methodology
Expected impact of the research results
Other:
19. If you have any more information to share about the motivation of content selection, please specify:
<Your answer>
20. What is the financial and/or sustainability model for the platform?*
Author pays
Optional purchases for authors (e.g. extra services like language editing)
Reader pays (e.g. freemium)
Grants to platform pay
Stakeholder consortium pays
Government pays
Host institution / platform owner pays
Partnering publisher pays
Voluntary donation
Business model not disclosed
Other:
21. Is your infrastructure based on:*
Open source software
Proprietary software
Other:
22. If you checked the box 'open source software', please provide a link to the source code if available:
<Your answer>
23. Does your infrastructure have *
Open APIs
Interoperability with other services
Other:
24. If you checked the box 'Interoperability with other services' please provide more information/examples:
<Your answer>
25. What does your platform offer regarding copyright/licensing?*
Open licensing - CC0
Open licensing - CC BY
Open licensing - CC BY-SA
Open licensing - CC NC
Open licensing - CC ND
Open licensing - CC NC-SA
Open licensing - CC NC-ND
Open licensing - GNU or other(s)
Other:
26. What does your platform offer regarding copyright?*
Creative Commons Attribution 4.0 International License (CC-BY 4.0)
