Developer Perceptions on Utilising Low-Code Approaches to Build Accessible and Adaptive Applications for Seniors

TL;DR

This paper evaluates a low-code tool, AdaptForge, through interviews with practitioners, highlighting its potential to simplify the creation of accessible, adaptive applications for seniors and guiding future tool design.

Contribution

It introduces an empirical evaluation of AdaptForge, a low-code MDE tool, and offers recommendations for designing accessible, adaptive software development tools for seniors.

Findings

Practitioners see AdaptForge as a promising industry-standard solution.

Automated code generation mitigates development constraints.

Recommendations improve future low-code tool designs for accessibility.

Abstract

The global ageing population presents a growing societal challenge, creating an urgent need for inclusive technologies that promote autonomy among older adults. Software practitioners can address this by delivering digital services that enhance seniors' independence and reduce reliance on routine support from family members and healthcare infrastructure. However, traditional development practices, constrained by time and resources, often result in applications with major accessibility and personalisation barriers. Increasing pressure from regulatory requirements, such as the European Accessibility Act (EAA), and the personal empathy many developers feel toward supporting their older loved ones and their own future selves have created a demand for tools that support the development of accessible and adaptive software. To address this demand, this paper presents an interview-based empirical study with 18 software practitioners, evaluating AdaptForge: a low-code model-driven engineering (MDE) tool that enables the efficient creation of accessible and adaptive applications for senior users by mitigating development constraints through automated code generation. Based on these insights, we identify developer expectations for adopting such tools as industry-standard solutions and provide empirically grounded recommendations for designing low-code tools that support accessible and adaptive software development.