Editor for accessible images in e-Learning platforms
Sandra Sanchez-Gordon, Juan Estevez, Sergio Luján-Mora · 2016 · Proceedings of the 13th International Web for All Conference (W4A) · doi:10.1145/2899475.2899513
Summary
This extended abstract defines a set of twenty features that an HTML visual text editor should provide to support content authors in creating accessible images within e-learning platforms. The features were derived from reviewing published research on accessible images and accessibility standards including WCAG 2.0, ATAG 2.0, ISO/IEC 24751, and IMS Access for All 3.0. The features span three areas: alternative text management (inserting, editing, help text, warnings for missing alt text, avoiding auto-generated alt text, retaining alt text in media libraries), long description support for complex images (inserting and editing longdesc URLs, contextual help, suggesting long descriptions when alt text exceeds 140 characters), and image title management. The authors used these twenty features as a rubric to assess eight e-learning platforms. They also prototyped an HTML visual text editor designed to be a reusable base component across platforms, featuring WCAG conformance check buttons (A, AA, AAA), verbose/advanced modes for contextual guidance, image type classification (decorative, simple, complex, link/button), and intelligent warnings for problematic alternative text patterns.
Key findings
The assessment of eight e-learning platforms revealed significant variation in image accessibility support. Moodle scored highest with 15 of 20 features, followed by Sakai (13), ATutor and MiriadaX (12 each), BrightSpace (9), Canvas (7), and edX and Udemy tied at the bottom with only 6 features each. An important finding was that platforms sharing the same underlying HTML editor component (Moodle/ATutor and Sakai/MiriadaX) showed similar but not identical scores, with differences attributable to version variations and customizations. This suggests that building a single, fully accessible base editor component could improve accessibility across multiple platforms simultaneously. The prototype editor includes practical intelligence: it warns about suspicious alt text (file names used as alt text, generic words like "photo" or "image of"), redundant alt text (duplicating nearby descriptions), and overly long alt text that may indicate a long description is needed. The image type selector prompts appropriate behavior—decorative images handled via CSS, complex images triggering long description fields, and link/button images redirecting to appropriate toolbar functions.
Relevance
This paper addresses a critical bottleneck in e-learning accessibility: content authoring tools that fail to guide educators in creating accessible images. Since instructors and students are the primary content creators in e-learning, the accessibility of the authoring tools directly determines the accessibility of educational content. The twenty-feature rubric provides a practical evaluation framework that platform developers and procurement officers can use to assess and compare e-learning platforms. The prototype editor's approach of intelligent warnings (rather than just validation) and contextual help that can be toggled between verbose and advanced modes reflects good understanding of author needs at different experience levels. The finding that no platform achieved all 20 features underscores the gap between accessibility standards and tool support. As a two-page extended abstract, the work lacks user evaluation with content authors, and the prototype had not yet been implemented in a production platform. The feature set also focuses exclusively on images, though the authors planned to extend to text, audio, and video accessibility features.
Tags: image accessibility · alternative text · e-learning accessibility · authoring tools · WCAG compliance · content creation · long description
Standards referenced: WCAG 2.0 · ATAG 2.0 · ISO/IEC 24751 · IMS AccessForAll