ITHACA: An Open Source Framework for Building Component-Based Augmentative and Alternative Communication Applications
Alexandros Pino, Georgios Kouroupetroglou · 2010 · ACM Transactions on Accessible Computing · doi:10.1145/1786774.1786775
Summary
This paper introduces ITHACA, an open source software framework for developing modular, customizable AAC applications. The authors address a persistent problem in assistive technology: AAC products are typically expensive, monolithic, difficult to customize, and limited in language support—making them inaccessible to many who need them. ITHACA proposes a fundamentally different approach by combining three software engineering paradigms: Component-Based Development (CBD), Open Source software development, and Design for All principles. The framework allows developers to create independent, interoperable software components (on-screen keyboards, symbol selection sets, text-to-speech engines, scanning input modules, chat/email clients) that can be mixed and matched by "integrators" (therapists, rehabilitation centers, families) to build personalized communication aids. Components communicate through a standardized protocol using "Interlingua"—a concept-based pseudo-language that enables translation between natural languages and symbolic communication systems like Blissymbols, PCS, and MAKATON. ITHACA introduces a hybrid Open Source/Commercial model where both free community-developed components and commercial components can coexist and interoperate. This addresses the economic reality of AT development: the market is too small for traditional commercial software economics, yet users need sustainable, professional-quality products. The framework runs on Windows using COM+ technology, chosen for its widespread adoption and developer familiarity, with all core components released under GPL.
Key findings
Developer evaluation involved 20 programmers over nine months building ten core components and eleven AAC modules. Quality metrics showed strong results: Functionality (4.39/5), Portability (4.35/5), Usability (4.31/5), and Maintainability (4.09/5). Code reuse was extensive, ranging from 40% to 88% across components—far higher than the predecessor closed-source ULYSSES framework. Open Source development also produced better-documented code (36% comments vs. 12% in closed-source), as developers knew their code would be publicly scrutinized and reused. End-user demonstrations with four children with cerebral palsy (ages 6-13) in Greece showed substantial communication improvements. One 6-year-old user expanded vocabulary from 10 PCS symbols to 99 concepts over three months—a 990% increase. All users showed vocabulary growth where therapists reported traditional methods had "reached a dead end." The personalized systems used different symbol sets (PCS, BLISS, MAKATON, natural Greek), input methods (five switches with directed scanning, touchscreen, puff switch, single switch with automatic scanning), and output modalities based on individual needs and abilities. Therapists, facilitators, and families reported that direct involvement in configuring and modifying the systems—enabled by the modular architecture—maintained user engagement and allowed rapid adaptation as needs changed. The ability to produce synthetic speech and correct errors was rated highly significant by educators.
Relevance
ITHACA represents an important model for addressing systemic problems in assistive technology markets. The combination of Open Source development, component modularity, and Design for All principles offers a template for making AT more affordable, customizable, and sustainable. The framework recognizes that AAC solutions require involvement of multiple stakeholders—developers, therapists, families, educators—and provides tools appropriate for each role. For accessibility practitioners, the paper demonstrates how software architecture decisions directly impact real-world accessibility outcomes. Monolithic commercial applications lock users into fixed functionality; component-based systems allow the personalization that people with complex communication needs require. The "Interlingua" concept—enabling interoperability between different symbolic languages and natural languages—addresses the multilingual challenge that limits many AAC products to English-speaking markets. The developer-evaluation methodology provides a model for assessing AT framework quality using ISO 9126-based metrics. The end-user demonstrations, while limited in formal rigor, illustrate the human impact of enabling ongoing customization: users who had plateaued with traditional approaches showed renewed progress. The participatory approach—involving families in development feedback loops—aligns with contemporary AT best practices emphasizing user involvement throughout the product lifecycle.
Tags: augmentative and alternative communication · AAC · open source · assistive technology · component-based development · Design for All · speech synthesis · symbol communication · cerebral palsy · software framework