The HTML output is designed to meet WCAG 2.1 Level AA, and the community takes this seriously. But compliance is ultimately a property of a specific document, not of a tool. If an author skips alt text, no toolchain can compensate.
Accessible PDFs are possible but genuinely difficult to produce well from LaTeX, and the result is often imperfect. HTML is fundamentally a more accessible format: it reflows, works natively with screen readers, and adapts to user preferences for font size and contrast. Generating HTML from the same PreTeXt source costs you nothing extra.
Yes, from the same source file. Formulas are rendered in Nemeth Code, and tactile diagrams can be generated automatically from structured figures. If you write mathematical images with PreFigure, that same source can produce the visual figure, a tactile diagram, and accessibility annotations for the image. The PreTeXt section on PreFigure shows how this fits into a PreTeXt project. The approach has been validated on full-length textbooks in abstract algebra and calculus, with quality checked by a certified transcriber and readability confirmed by a blind mathematician.
Because mathematicians read math by looking back and forth, comparing pieces, holding sub-expressions in view. Braille gives a blind reader the same tactile access.
No. The authoring practices in What You, the Author, Must Do are your primary responsibilities. You don’t need to study WCAG to produce accessible output.
Ultimately your institution and your students. PreTeXt gives you a document that is accessible by default and easy to improve where it matters most. For specific student accommodations, work with your campus disability services. The HTML, EPUB, and Braille outputs will cover most common needs out of the box.
Every non-decorative image needs a <shortdescription>: plain text, no markup, no quotation marks, under about 125 characters. Complex images can also include a <description> with paragraphs and even math, which screen readers read as a longer alternative.
Because image descriptions depend on author intent and context. The same graph might be described one way in a section about asymptotes and another way in a section about symmetry. No tool can reliably make that call.
Don’t rely on color alone. If you ask students to compare the red and green curves, add a second distinguishing feature. For example, dashing, thickness, or labels. A quick test: view your image in black and white and check that it still conveys the intended information.
No. Producing Braille involves additional tools beyond the setup we’re using. The capability exists and the HTML you build today is already screen-reader accessible. Braille is a next step for instructors who need it.
Yes. Many US institutions are now required to ensure accessibility of all online materials, including content posted in an LMS like Canvas or Blackboard. PreTeXt is well-positioned for this: you can deploy the HTML publicly, or export a single portable HTML file that uploads directly to an LMS.
Native PreTeXt exercise types (multiple choice, true/false, fill-in) are generally accessible. Embedded third-party interactives (GeoGebra, Desmos, DoenetML) inherit the accessibility of those platforms, which varies. For critical content, consider including an image or text description as a fallback.
