Empathic communication of accessibility barriers in web 2.0 editing
Afra Pascual, Mireia Ribera, Toni Granollers · 2015 · Proceedings of the 12th International Web for All Conference (W4A) · doi:10.1145/2745555.2746642
Summary
This paper presents EmpathicEditor4Accessibility (EE4A), a prototype web editor that helps non-technical content writers understand and fix accessibility errors by communicating barriers through the perspective of disabled personas rather than technical jargon. The authors identify a critical gap: the explosion of user-generated content means millions of non-technical "prosumers" are publishing web content through CMS platforms without awareness of accessibility, yet existing evaluation tools communicate errors using WCAG success criteria numbers, HTML code references, and technical language that even developers find hard to understand. EE4A takes a semiotic engineering approach, framing accessibility errors as communication between disabled end users and content writers. The tool was built around detailed personas — such as "Teresa Sinvision," a 29-year-old blind telephone operator — created from real testing sessions where people with cognitive, visual, auditory, and motor disabilities interacted with intentionally inaccessible content. Error messages are written from the user's perspective (e.g., "Teresa is visually impaired and needs someone to explain the content of the picture to her" instead of "img element missing alt attribute"). The editor also integrates disability simulation tools showing how content appears to users with blindness (screen reader view), low vision (magnifier with high contrast), and colour-blindness.
Key findings
A comparative test with 8 prosumer users (bloggers and institutional website editors) showed substantial improvements over TAWCMS, a traditional accessibility evaluation tool. TAWCMS took 26-45% longer across three tasks (alt text, link descriptions, heading hierarchy). More strikingly, correct error interpretation jumped from 50%/25%/37.5% with TAWCMS to 100%/75%/100% with EE4A across the three tasks. All 8 users preferred EE4A, with comments like "I felt much less sure and more nervous when trying to understand the errors shown in TAWCMS." Eye tracker analysis revealed that within the same 15-second window, TAWCMS users were still reading the error explanation, while EE4A users had already moved on to the editing area to fix the problem. Users without HTML knowledge particularly valued the non-technical language. Even users with HTML experience noted that TAWCMS was "excessively technical, even for users with a technical profile." A notable collateral finding was that users found it genuinely difficult to write image alt text, suggesting this is not a trivial task even when people understand why it matters.
Relevance
This research directly addresses one of the biggest bottlenecks in web accessibility: the gap between technical accessibility standards and the non-technical people who actually create most web content. The approach of using personas and empathic language to communicate accessibility errors is immediately applicable to any CMS or authoring tool. The finding that even HTML-knowledgeable users found traditional error messages excessively technical reinforces that the communication problem is not just about technical literacy — it is about framing accessibility in human terms that motivate action. The persona-based approach also implements ATAG principles that remain underadopted in commercial authoring tools, demonstrating a practical path toward the ATAG vision of tools that actively support authors in creating accessible content. The difficulty users experienced writing alt text points to a need for better guidance and examples within authoring interfaces, not just error detection.
Tags: authoring tools · user-generated content · accessibility education · empathic design · WCAG compliance · content management systems · personas
Standards referenced: WCAG 2.0 · ATAG 2.0 · UAAG 2.0