← All reviews

A4TV: Assessing and Ameliorating the Accessibility of the Ascending Connected TV Platforms

Daniel Costa · 2015 · Proceedings of the 12th International Web for All Conference (W4A) · doi:10.1145/2745555.2746669

Summary

This doctoral consortium paper proposes a two-stage research programme to study and improve the accessibility of connected TV platforms for visually impaired users. Connected TVs — smart TVs and set-top boxes with internet connectivity — use web technologies (HTML5, JavaScript) running in browser-based runtime environments, inheriting both the accessibility barriers and solutions found on desktop web. However, the TV environment introduces distinct constraints: different screen distances, limited space for on-screen information, different navigation methods (remote controls rather than keyboards), and a near-complete absence of assistive technology providing audio feedback for what appears on screen. The author argues that the transition from analogue to digital TV has paradoxically made television less accessible to blind users — while analogue TV was simple (change channels, adjust volume), digital TV demands navigation through complex on-screen menus, electronic programme guides, and web applications that are inherently visual. Prior research on TV accessibility (the GUIDE project, Epelde et al.) focused primarily on elderly users, leaving blind users' interaction with connected TV applications almost entirely unexplored.

Key findings

The first stage combined a survey of visually impaired users, automated WCAG 2.0 conformance testing, and a user study comparing TV and desktop versions of the same web applications. The survey confirmed that blind users have significant difficulties with standard TV features beyond channel switching — accessing programme guides, recording shows, and other interactive features — and would welcome using their TV for content access if it were made more accessible. A counterintuitive finding from the automated evaluation was that TV versions of web applications actually showed higher WCAG conformance than their desktop counterparts, likely because TV interfaces are simpler and contain less content. However, the user study revealed that these automated conformance results did not translate to better real-world usability, reinforcing existing research (Vigo et al., 2013) that automated evaluation alone is insufficient. The second stage proposes an "Accessible Smart Remote Control" concept: distributing the TV interface across paired TV and mobile devices, using the mobile phone as an accessible control surface that can provide screen reading, content adaptation, and multimodal interaction. The approach leverages WAI-ARIA to expose TV application semantics to assistive technologies on the mobile device.

Relevance

This paper identifies an important and growing accessibility gap: as television becomes an interactive, web-connected platform, it becomes simultaneously less accessible to blind users who previously managed analogue TV independently. The finding that automated WCAG testing gives misleadingly positive results for TV applications — while actual user experience remains poor — is a valuable cautionary example for the accessibility testing community. The second-screen approach (using a mobile phone as an accessible companion to the TV) is pragmatic and forward-looking, building on devices and interaction patterns users already have rather than requiring TV manufacturers to build in assistive features. For practitioners, the paper highlights that accessibility standards designed for desktop web do not straightforwardly transfer to TV contexts, and that platform-specific evaluation and design are needed. However, this is a 2-page doctoral plan rather than a completed study, and the second-stage prototypes are described in concept only.

Tags: connected TV · blindness · visual impairment · accessibility evaluation · multimodal interaction · WAI-ARIA · WCAG · mobile accessibility · second screen · audio description · automated testing

Standards referenced: WCAG 2.0 · WAI-ARIA 1.0