Crosschecking the Mobile Web for People with Visual Impairments
Luís Carriço, Rui Lopes, Rogério Bandeira · 2011 · Proceedings of the International Cross-Disciplinary Conference on Web Accessibility (W4A) · doi:10.1145/1969289.1969305
Summary
This paper proposes a framework for evaluating mobile web accessibility for specific disability profiles by coherently merging three sets of guidelines: WCAG (web accessibility), W3C Mobile Web Best Practices (MWBP), and disability classifications from the ICF. The authors argue that simply combining WCAG and MWBP is insufficient because the relationships between the three dimensions — accessibility guidelines, mobile best practices, and specific disabilities — are not straightforward. A key insight is that excluding a WCAG guideline for a specific disability does not automatically make the corresponding MWBP guideline irrelevant, and vice versa. For example, WCAG's color use guideline might be irrelevant for a non-visual impairment on desktop, but the corresponding MWBP check still applies on mobile because devices may have poor color contrast. Conversely, some MWBP guidelines become irrelevant for specific disabilities based on actual usage patterns — if a blind user disables image loading in their browser, then MWBP checks for image sizing, resizing, and large graphics become moot. The paper proposes a three-step evaluation rationale: (1) filter WCAG by disability type, (2) filter MWBP by disability type and usage patterns, (3) apply the WCAG/MWBP effort-level mapping only to the remaining subsets.
Key findings
A questionnaire study with 18 Portuguese-speaking visually impaired users (8 blind, 9 partially sighted, ages 18-60) investigated browser configuration patterns. All were daily desktop web users; half browsed the web on mobile at least weekly. Key findings: blind mobile web users largely confirmed the hypothesized pattern — 3 of 5 blind mobile users disabled image loading, and a 4th did so on desktop and complained about mobile page loading speed. Partially sighted users did not confirm the pattern — none disabled image loading on either platform, even though they used magnification and font enlargement. This distinction is significant: the authors argue that on small mobile displays, low-vision impairment is further stressed, and low-vision users effectively cannot access mobile web content without screen reader capabilities. The study also revealed that performance concerns (page loading speed) were a key reason blind users avoided mobile web browsing, and disabling images helped address this. The paper presents disability-specific mapping tables showing which WCAG checkpoints and MWBP guidelines are relevant for blind, low-vision, deaf, color-blind, motor impaired, and cognitive disability profiles.
Relevance
This paper tackles the often-overlooked intersection of mobile web quality and disability-specific accessibility — two domains with their own guideline sets that interact in complex ways. The framework's core contribution is the insight that accessibility evaluation should be context-aware: the same guideline may be critical in one context (desktop + low vision) and irrelevant in another (mobile + blind user with images disabled). This anticipates modern responsive accessibility thinking where evaluation must consider device, user configuration, and AT together. The finding that blind and low-vision users have fundamentally different mobile browser configuration patterns challenges the common practice of treating "visual impairment" as a single evaluation category. The small sample size (18 participants) limits generalizability, and the 2011 mobile landscape has changed dramatically — modern smartphones with VoiceOver/TalkBack have made mobile web browsing far more viable for blind users. However, the underlying principle that evaluation frameworks must account for actual user behavior and device configuration, not just abstract guideline compliance, remains sound.
Tags: mobile accessibility · blind and low vision · WCAG compliance · accessibility evaluation · web standards · user research
Standards referenced: WCAG 1.0 · WCAG 2.0 · MWBP · ICF