In September 2026, the OpenAlex Colors Working Group, convened by the OpenAlex Community Advisory Board, published a document titled Recommendations for Revising the Open Access Classification System in OpenAlex (Alperin et al., 2026), along with an explanatory blog post, provocatively entitled Is it time to retire the OA colours? Tell us what you think (Demes, 2026). In a nutshell, it proposes to replace the current journal-based colour schema for open access (OA) categories (gold, green, diamond, bronze) with a three-dimension classification (access, location, license) at the document level (article, dataset, book, proceeding). At Érudit, we welcome the opportunity to re-examine these well-established categories and debate on whether they are still able to account for the many traits of open access. However, any suggestion for change must reflect on what would be gained against what would be lost in the process.
In 2022, OpenAlex emerged as one of the most relevant initiatives for open scientific information, eliciting great expectations from the bibliometrics, metaresearch, research assessment, and scholarly publishing communities, and eventually becoming an indispensable tool. Open infrastructures such as OpenAlex are key in the visibility, preservation and valorization of research marginalized by commercial databases and indexes. Such research is, in turn, often found in diamond journals. More than a business model, in the diamond approach, conditions of production and of access are not separate dimensions: how an article is produced and how it gets to be accessible are intrinsically linked. Classifying a journal as diamond denotes not only a no-fee-for-authors-nor-readers schema, most importantly, it highlights scholarly leadership, community ownership, and public support for knowledge as a common good (Ancion et al., 2026). These journals are collective editorial projects and not a collection of individual articles. This is why we believe that OpenAlex’s proposition, while potentially achieving more precise data granularity at the document (‘work’) level, presents shortcomings that will weaken the position of diamond journals in an already uneven global playing field.
OpenAlex’s proposal of re-defining openness by creating the three dimensions ‘access’, ‘location’ and ‘licence’, although elegant at first sight, are themselves a manifestation of particular, longstanding logic of commercial models wherein articles are treated as alienable commodities, free to be sold or shared in a market. Focussing on access, location and licence replicates a limited understanding of open access. In support of forms of scholarly exchange in the Humanities and Social Sciences (HSS), where “the journal becomes a focal point for the community [of scholars] itself” (Frantsvåg, 2026), Érudit’s position is that the introduction of these new dimensions should be accompanied by the development of practical journal-level metadata.
We anticipate the proposed shift of focus towards works will lead to a significant loss of knowledge (and, therefore, usefulness) for a range of user communities. While OpenAlex has been adopted widely for article-level bibliometrics, many users require journal-level information. For example, hybrid open access article processing charges are not admissible for cOAlition S funders, yet the proposed changes will prevent funders from distinguishing gold (eligible) and hybrid (ineligible) journals.
Even more journals will be concerned by the disappearance of reference to diamond, an open access category which, regrettably, is not mentioned at all in the Recommendations proposal, nor in the conversion table of Appendix B, although it has been presented by OpenAlex for almost a year now. Diamond open access is at the core of many national policies and not only should it remain visible in OpenAlex’s metadata, it should also be straightforward to extract this information for both advanced users and novices. While we appreciate the idea of a separate business model dataset, we stress the importance of linking article-level and journal-level metadata. Specifically, we recommend the development of practical journal-level metadata documenting fees for readers and authors, i.e. the usage of article processing charges by a journal and the presence of a reading paywall.
In addition, the distinction of the work’s source, ‘publisher’ or ‘platform’, is ambiguous and creates confusion for users. The recommendations report proposes ‘publisher’ being attributed to venues “whose primary function is to publish scholarly works”, while ‘platforms’ would be defined as “a repository, preprint server, or other hosting service”. However, multiple independent journals are not linked to a publisher, have no website of their own, and are solely disseminated by platforms whose primary function is to publish scholarly works, such as J-STAGE, SciELO, HRČAK, African Journals Online, and our platform, Érudit.
Open science projects like OpenAlex should seize the opportunity to better reflect on how distinctive journals and disciplines publish. This proposed solution risks entrenching dominant visions that cleave questions of access from those of production and devalue publishing norms for communities supporting the diamond model, in the Global South as well as the Global North. By considering access only at the endpoint, i.e. for reading or re-using, it endorses a shallow definition of openness that does not include conditions of production of open knowledge. This approach also marginalizes regional and national efforts that support diamond open access and continue to invest in non-commercial publishing systems.
Researchers and publishing communities, as well as universities and funders, need a common nomenclature for publishing models. The classification of gold, green, hybrid and diamond is not artificial; it is a result of historical developments of the scholarly publication ecosystem. Although many admit to the current terms’ ambiguities, they remain highly useful in communicating policies and research results and they have been implemented widely. Érudit hopes OpenAlex continues to grow and become the main reference in metaresearch; yet we urge OpenAlex to consider the interests of the communities engaged in open, diverse, and equitable publication practices, by maintaining the inclusion of metadata that highlight its distinctive features: those that define publishing models.
References
Alperin, J. P., Kramer, B., Farley, A., Church, A., van Eck, N. J., Mitchell, D., Marcum, C., Takenaka, R., Harrison, M., & Gatti, R. (2026). Recommendations for Revising the Open Access Classification System in OpenAlex. OpenAlex Community Advisory Board. https://doi.org/10.5281/zenodo.21934115
Ancion, Z., Arbain, A., Babini, D., Ball, J., Becerril-García, A., Borrell-Damián, L., Deschênes, M., Estrada-Medina, M. A., Farley, A., Gutam, S., Oliva Uribe, D., Rooryck, J., Saenen, B., Varoglu, Z., & Gungula, E. W. (2026). Bengaluru Roadmap and Action Plan on Diamond Open Access. International Overseeing Committee, 3rd Global Summit on Diamond Open Access. https://doi.org/10.5281/ZENODO.20168763
Demes, K. (2026, September 17). Is it time to retire the OA colours? Tell us what you think. OpenAlex Blog. https://blog.openalex.org/is-it-time-to-retire-the-oa-colours-tell-us-what-you-think/
Frantsvåg, J. E. (2026, 27 October). Paradise Lost, Knowledge Regained: Jan Erik Frantsvåg on reclaiming open access. DOAJ Blog. https://blog.doaj.org/2026/10/27/paradise-lost-knowledge-regained-jan-erik-frantsvag-on-reclaiming-open-access-part-2/