arXiv Launches HTML Papers for Enhanced Accessibility

arXiv is committed to enhancing accessibility in research dissemination by introducing HTML paper formats alongside the traditional PDF. Recognizing the urgent need expressed by our community, this initiative aims to make scholarly content more accessible.

We have successfully launched HTML papers and are progressively converting arXiv's extensive corpus of over 2 million papers. While most papers will be successfully converted, a small percentage may not have an HTML version; we are continually working to improve our conversion process. Authors will have the opportunity to preview their paper's HTML during the submission process, and links to the HTML format will be available on abstract pages, beneath the existing PDF download option.

This beta rollout marks just the beginning of our efforts. We are dedicated to continuously improving HTML papers and will actively seek feedback from authors, readers, and the entire arXiv community to refine conversions from LaTeX.

Why "Experimental" HTML?

Approximately 90% of arXiv submissions are in TeX format, primarily LaTeX. This presents a unique accessibility challenge: accurately converting from TeX—an highly extensible language used in diverse ways by authors—to HTML, a format significantly more compatible with screen readers, text-to-speech software, screen magnifiers, and mobile devices. Beyond the technical complexities, the conversion must be both rapid and automated to uphold arXiv's core service of free and swift dissemination.

Given these challenges, we anticipate some conversion and rendering issues. We have opted for an "experimental" HTML beta launch because:

• Immediate Accessibility: The arXiv community, particularly researchers with accessibility needs, has strongly advocated for immediate action.
• Community Collaboration: While significant work has been completed, community reports are crucial for identifying issues linked to specific LaTeX packages that are not converting correctly.

Understanding HTML Paper Errors

HTML papers on arXiv.org are a continuous work in progress, and occasionally, errors may appear. As we strive to enhance accessibility, we aim to clarify the causes of these errors and suggest how authors can help minimize them.

How You Can Help

1. Read HTML Papers and Report Issues:

   We encourage the community to explore HTML papers within their fields. To report an issue:

   • Navigate to the abstract page of a paper.
   • Locate the HTML link below the PDF download link and click it.
   • Report issues using one of the following methods:
     • Click the "Open Issue" button.
     • Select text and click the "Open Issue for Selection" button.
     • Use Ctrl+? on your keyboard. Screen reader users can use Alt+y to toggle accessible reporting buttons per paragraph.

   It's important to understand that our primary goal is accessibility, valuing function over form during this beta phase. Therefore, please refrain from reporting that an HTML paper does not look exactly like its PDF counterpart. While incorrect or illegible HTML layouts are critical to report, we expect stylistic differences. Line breaks will vary, and there will likely be more whitespace, resulting in a less compact presentation. Intricate typographic layouts will not be rendered with the same level of detail, by design. HTML, as a distinct medium, offers unique advantages over PDF, including superior compatibility with assistive technologies and better adaptability to various reading devices, including mobile.

2. Help Improve LaTeX Conversion:

   • Authors: You can contribute by adhering to our guide on LaTeX Markup Best Practices for Successful HTML Papers.
   • Developers: If you have available development cycles, your contributions are welcome! Our collaborators at LaTeXML maintain a list of issues and actively seek feedback and developer contributions.
   • Publishers, Society Members, Conference Organizers: You can assist by reviewing the .cls files your organization recommends to authors for unsupported packages. Promoting .cls files that utilize supported packages is a simple yet impactful way to foster accessibility in the scientific community.

Acknowledgments

We extend our sincere gratitude to all scientists with disabilities who have generously shared their insights, expertise, and guidance throughout this project. We also thank two pivotal organizations whose contributions made HTML papers on arXiv possible: The LaTeX Project and the LaTeXML team from NIST. Their knowledge, incredible work, and unwavering commitment to accessibility are deeply appreciated.