Knowledge Barrier-Less Press All articles
Health & Medicine

Alt Text Is Not a Luxury: The Accessibility Debt Inside Open Science

Knowledge Barrier-Less Press
Alt Text Is Not a Luxury: The Accessibility Debt Inside Open Science

Photo: Tomascastelazo, CC BY-SA 3.0, via Wikimedia Commons

Open access was supposed to remove the tollbooth from scientific reading. It did not remove the staircase. A paper can be free to download and still be unreadable to a blind chemist using a screen reader, a dyslexic undergraduate using text-to-speech, or a reader who depends on high-contrast display and keyboard navigation. The file is open. The knowledge is not.

This is the accessibility debt of the open-science movement. Publishers, universities, funders, and authors have spent years arguing about who should pay for access. Far less attention has gone to whether the resulting articles can actually be used by people with disabilities. Too many scholarly PDFs are scanned images wrapped in a thin layer of text. Too many figures have no alt text. Too many charts rely on color alone. Too many equations are embedded as pictures. Too many platforms fail basic keyboard navigation.

A PDF Is Not a Format. It Is a Barrier.

For a sighted reader with a fast laptop, a PDF feels neutral. For a blind reader, it can be a maze. A tagged PDF can work with a screen reader if it is built correctly. An untagged PDF often reads as a stream of disconnected characters or skips entire sections. A scanned page may be searchable but still inaccessible. A figure may be essential to the argument and yet exist only as an image with no description.

Dyslexic readers face a different set of obstacles. They may need adjustable line spacing, clean sans-serif fonts, shorter line lengths, and text that can be read aloud without losing mathematical notation or table structure. Readers with low vision may need to zoom to 400 percent without the page collapsing into overlapping columns. Readers with motor disabilities may need to navigate by keyboard rather than mouse. Readers with cognitive disabilities may need clear headings and plain-language summaries.

The common thread is design. Accessibility is not a special feature bolted onto a finished product. It is a set of decisions made at the beginning: how to structure a document, how to describe a figure, how to mark up a table, how to caption a video, how to test a platform with real users.

The Legal Floor Is Not the Ceiling

In the United States, the legal landscape is clearer than it once was. The Americans with Disabilities Act prohibits discrimination on the basis of disability. Section 508 requires federal agencies to procure and use accessible information and communication technology. The Department of Justice's 2024 rule on web accessibility for state and local governments put public institutions on notice that digital services must meet recognized standards. The widely used Web Content Accessibility Guidelines, now at version 2.2, provide a technical benchmark.

But compliance is not the same as usability. A publisher can claim that a platform meets a checklist while still producing articles that are miserable to navigate. A university can buy an accessible learning-management system while faculty upload inaccessible PDFs. A funder can require data sharing while ignoring whether the shared data are described in ways that disabled researchers can use.

The scholarly communication system needs a higher standard: accessible by design. That means structured HTML as the primary format, not a PDF afterthought. It means MathML for equations, alt text for every meaningful figure, long descriptions for complex charts, properly marked-up tables, transcripts for audio and video, and metadata that tells assistive technology what it is reading. It means testing with blind, low-vision, dyslexic, and motor-impaired users, not just running an automated checker.

What Publishers, Universities, and Authors Can Do

Publishers should make accessible HTML the default and treat PDFs as supplementary files. They should require alt text at submission, provide author guidance, and employ accessibility specialists who can remediate complex figures. They should publish accessibility statements that are specific and honest, not vague promises.

Universities should treat accessibility as part of research infrastructure. Libraries and presses can offer remediation services, training, and procurement standards. Faculty should not be left to guess whether their course readings or journal submissions are usable. Graduate students and early-career researchers need accessible authoring tools and clear expectations from the start.

Authors can do a great deal without becoming accessibility engineers. Write alt text that explains what a figure shows and why it matters. Do not rely on red-versus-green distinctions alone. Use readable fonts and clear headings. Provide a plain-language summary. Caption videos. Share data in accessible formats. Ask journal editors whether the final article will be tested with assistive technology.

Funders should go further. Federal agencies that spend billions on research can require accessibility plans in grants and verify that published outputs meet recognized standards. The National Institutes of Health and the National Science Foundation already shape how research is conducted and shared. They can shape whether disabled readers are included in the audience for that research.

Nothing About Us Without Us

Accessibility is often treated as a technical specialty, but it is also a question of who science is for. If open science means knowledge without barriers, then a barrier that keeps out blind and dyslexic readers is not a minor defect. It is a contradiction at the center of the project.

The good news is that the fixes are known. They are not mysterious. They require money, training, standards, and humility. They require publishers to stop treating accessibility as an expense and start treating it as a condition of publication. They require universities to stop assuming that free means open. And they require the rest of us to remember that a scientific article is not truly public until the public can read it.

All Articles

Related Articles

The Badge Tax: Why the Most Important Science Talks Are Still Behind a Registration Desk

The Badge Tax: Why the Most Important Science Talks Are Still Behind a Registration Desk

Free to Read, Impossible to Understand: Why Open Access Hasn't Kept Its Promise to the American Public

Coded Diagnosis: The Invisible Algorithms Making Medical Decisions You Never Agreed To

Coded Diagnosis: The Invisible Algorithms Making Medical Decisions You Never Agreed To