Skip to main content
← Back to Home

Accessibility

I want this site to be usable by as many people as possible. I use WCAG 2.2 Level AA as the guide for that work.

Current status

I have completed a substantial, evidence-backed accessibility review and remediation program across the current website, preserved historical pages, rescued archive samples, locally published documents and media, and the separately generated CV.

As part of that work, an internal evaluation on August 18, 2026 covered the 1,372 HTML pages then generated by this website. After remediation, it found no known WCAG 2.2 Level A or AA blockers in that defined scope. The page population changes as content changes and continues to receive automated checks. This is owner-led evidence, not third-party certification or a claim that every file on the domain conforms.

A September 9 follow-up restored the reviewed, same-origin images in 72 active 2004–2005 NQB2 archive posts and replaced seven third-party image sources with accessible text or link alternatives. A further bounded layout review restored seven inline icons and three image floats in historical posts, with stacked layouts on narrow screens, and removed six unavailable decorative icons while retaining their labelled links. Larger unclassified image placements remain for separate review.

A September 13 follow-up restores seven early NQB photo pages with structured headings, descriptive image alternatives and layouts that adapt to narrow screens. These pages receive targeted automated and keyboard checks; their addition is not a new archive-wide conformance evaluation.

What the work includes

Every staging and production build checks the structure of every HTML page generated from the website source. Every proposed change that can affect rendered pages also runs a fixed browser-based accessibility gate over representative pages and interactions. Documentation-only changes retain the required build check without provisioning a browser, while a recurring automated crawl checks the complete generated-page population. I can run that complete crawl on demand before especially consequential changes.

Because automation covers only part of accessibility, the program also includes representative keyboard, zoom and reflow, media, VoiceOver, ChromeVox, and browser review. It includes bounded review and repair of preserved historical pages and rescued archive content, a document-and-media inventory, and separate checks for the generated CV HTML and tagged PDF.

The local search added after the August evaluation has separate Chromium, Firefox, and WebKit checks for keyboard use, filters, no-JavaScript navigation, and failure handling. It also received owner-operated VoiceOver, zoom and reflow, and touch review before publication.

Known limitations

Testing cannot cover every page, file, device, browser, or assistive technology. This review was conducted by the site owner and did not include an independent specialist audit or a formal evaluation with disabled users. Four historical music PDFs now have structured HTML alternatives, but the original PDFs and other older documents, media, and preserved pages may still present barriers. Five owner-reviewed historical PDFs restored in September 2026 preserve their original, untagged document bytes and may also present barriers; no equivalent HTML alternative is claimed. The historical Grey Album MP3 downloads do not have transcripts or other text alternatives. The restored Berkeley talks have machine-generated HTML transcripts with timestamp links and occasional notes for possible untranscribed intervals. A second model compared the text with the audio and supplied exact high-confidence corrections, but only a small pilot set received individual human listening review. Each page explains that it has not received comprehensive human review. The remaining text may mishear names or words, omit speech, or merge questions with answers. Speaker changes and meaningful non-speech sounds are not identified; the original audio remains authoritative. Some historical music excerpts and the Tales from the Long Tail podcast also lack complete text alternatives. Local search requires JavaScript and WebAssembly; if a browser security mode such as Apple Lockdown Mode prevents it from starting, the search page instead keeps links to the complete writing archive and curated history available. Third-party services and external sites remain outside my control. If you find a barrier or something this review missed, please let me know using the contact information below.

Feedback and improvements

Accessibility is ongoing, and I welcome feedback and suggestions for improvement. If something on this site is difficult to use, please email me at joehall@gmail.com. You do not need to know the relevant standard or diagnose the cause. If you are comfortable sharing them, the affected URL, what you were trying to do, what happened, and your browser or assistive technology can help me investigate. Please do not include sensitive information.

Last reviewed: .