Open Access Science Publishing: What It Is and How It Works
Open access publishing has quietly reshaped how scientific knowledge moves through the world — who gets to read it, who pays for it, and what obligations researchers have to funders and the public. This page covers the definition of open access, the mechanics of its major models, the situations where it most commonly applies, and the practical boundaries that determine which path a researcher or institution chooses.
Definition and scope
A scientific paper published behind a paywall costs nothing to read if a university library has negotiated access — but that same paper is invisible to a physician in rural Ghana, a high school teacher in rural Kansas, or an independent researcher without institutional affiliation. Open access is the principle that peer-reviewed scientific literature should be freely available online, without subscription fees or access barriers, to anyone with an internet connection.
The Budapest Open Access Initiative, published in 2002, gave the movement its foundational definition: literature that users can "read, download, copy, distribute, print, search, or link to the full texts." That document drew a sharp line between access and ownership of copyright, a distinction that still governs how open access licenses are written today. The most widely used license framework is the Creative Commons Attribution (CC BY) license, which permits any reuse as long as the original authors are credited.
Scope-wise, open access applies across all scientific disciplines — from genomics to climate science — and is increasingly mandated by public funders. The National Institutes of Health (NIH) requires that any research it funds be made publicly accessible through PubMed Central within 12 months of publication. The National Science Foundation (NSF) updated its public access policy in 2023 to eliminate that 12-month embargo entirely for NSF-funded work. For deeper background on how scientific research is structured and validated, the conceptual overview of how science works provides useful grounding.
How it works
Open access is not a single publishing model — it is a spectrum of them. The three most consequential are Gold, Green, and Diamond.
-
Gold open access: The final published version of record is immediately and permanently free to read on the publisher's website. The cost is typically borne by the author (or their institution or funder) through an Article Processing Charge (APC). APCs at major journals range from roughly $1,000 to over $11,000 (PLOS, publisher APC schedule).
-
Green open access: The author deposits a version of the paper — often the accepted manuscript, not the final typeset version — into a publicly accessible repository, such as PubMed Central or an institutional repository, sometimes after a publisher-imposed embargo period.
-
Diamond open access: Neither authors nor readers pay. The publishing costs are absorbed by institutions, academic societies, or government grants. Diamond journals are common in mathematics and humanities but exist across sciences.
The most prominent journals and publishers now also operate hybrid models — subscription journals that allow individual articles to be made open access for a fee. Critics at the Scholarly Publishing and Academic Resources Coalition (SPARC) have noted that hybrid arrangements can result in institutions paying twice: once through subscription fees and again through APCs.
Common scenarios
Open access most frequently surfaces in three distinct situations.
Funded research mandates. When a study is funded by NIH, NSF, the Wellcome Trust, or any of the signatories to Plan S, the grant conditions typically require open access deposit or publication. Researchers who do not comply risk losing future funding eligibility.
Preprint servers. Before formal peer review, researchers post manuscripts to servers like arXiv (physics, mathematics, computer science) or bioRxiv (biology). These are not peer reviewed but are freely accessible and increasingly cited — a middle layer between private research and formal publication that became especially visible during the COVID-19 pandemic, when preprints moved faster than journal timelines.
Institutional repositories. Universities like MIT and Harvard maintain their own open access repositories and have adopted faculty mandates requiring deposit of scholarly work. MIT's open access policy, adopted in 2009, was among the first such institutional mandates in the United States.
Decision boundaries
Choosing between open access routes involves trade-offs that are financial, legal, and reputational.
The central financial question is who pays the APC. Many institutions have established agreements called transformative agreements with publishers — deals that bundle subscription and open access publishing costs — but coverage varies widely by institution and journal.
The legal question involves copyright retention. In traditional subscription publishing, authors typically transfer copyright to the publisher. Gold open access under CC BY allows authors to retain copyright. Green open access often imposes restrictions on which version can be shared and when.
The reputational question is more informal but real: some high-impact journals remain subscription-based, and the pressure to publish in them can conflict with funder mandates. The emergence of fully open access journals with strong impact factors — including PLOS Biology and eLife — has softened this tension, but it has not eliminated it.
The Science Journals and Publications section of this site covers the landscape of specific journals across disciplines. For the broader context of how scientific findings get funded and verified before reaching publication, The Science Authority's main reference maps the full knowledge base.