ASL-STEM Forum: Enabling Sign Language to Grow Through Online Collaboration
Anna C. Cavender, Daniel S. Otero, Jeffrey P. Bigham, Richard E. Ladner · 2010 · CHI '10: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems · doi:10.1145/1753326.1753642
Summary
This paper presents the ASL-STEM Forum, the first online video-based collaboration tool designed to help expand American Sign Language (ASL) vocabulary for Science, Technology, Engineering, and Mathematics (STEM) fields. The paper addresses a significant educational barrier: ASL lacks agreed-upon signs for many complex scientific and technical terms, causing deaf students to miss or misunderstand course material. When signs do exist for a term, they are often developed locally or regionally, leading to inconsistent usage that strains collaboration between deaf scientists and students at different institutions. While expert committees and video dictionaries have attempted to formalize STEM signs, the authors argue that natural language evolution occurs through community consensus, not by committee decree — and the small, geographically dispersed population of deaf STEM professionals makes organic vocabulary growth difficult. The ASL-STEM Forum addresses this by providing a community-based, video-enabled web resource organized hierarchically under four root topics (Science, Technology, Engineering, Mathematics). Each topic page displays an English term and definition, allows users to upload video sign suggestions via an integrated QuickCapture webcam applet, and supports discussion through text comments. Users can rate signs on a five-star scale, with the highest-rated sign displayed prominently. The system was built with Ruby on Rails, with user-created video hosted on YouTube. A motivating analysis found that of 657 new words added to the Oxford English Dictionary in 2008, 450 (69%) had already appeared in community-driven resources like Wikipedia, Wiktionary, or Urban Dictionary — on average 1.5 to 3.3 years earlier — suggesting that community-driven forums can identify and standardize vocabulary faster than formal processes.
Key findings
A 10-day user study with 14 ASL users involved in STEM fields (average age 36.3, geographically dispersed across the U.S., with lifelong sign language exposure averaging 26.1 years) showed the Forum's viability. Participants contributed a total of 106 video signs, 24 text comments, 18 ratings, and 9 new topics — 163 combined contributions. Contributing a sign took an average of just 2.27 minutes (SD=1.38), with the fastest at 0.78 minutes, demonstrating that the streamlined QuickCapture process successfully lowered the barrier to participation. Participants rated the Forum highly: 4.0/5 for ease of contribution, 4.8/5 for being a valuable resource, and 3.7/5 for ease of access. The study revealed three distinct levels of user participation mirroring typical online community patterns: advanced contributors who submitted many signs and comments, mid-level contributors with fewer submissions, and lurkers who primarily rated and browsed. Notably, the six most active sign contributors also commented the most and indicated in surveys that commenting was more valuable than numeric ratings, since a low rating alone doesn't explain whether a sign is inaccurate, unclear, or simply regionally unfamiliar. Participants were enthusiastic, with one stating "There is no doubt that this will be beneficial for the entire academic Deaf community!" and another noting the YouTube integration would "draw more Deaf people or hearing signers/interpreters to your site."
Relevance
This paper addresses a unique and underappreciated dimension of accessibility in education: the linguistic infrastructure needed for deaf students to access advanced academic content. While accommodations like interpreters and captioners are commonly discussed, they can only be as effective as the underlying sign language vocabulary allows. When a STEM concept has no established sign, interpreters must fingerspell the term (slow and error-prone for complex terminology) or improvise, leading to inconsistency across classes and institutions. For accessibility practitioners, the ASL-STEM Forum demonstrates how technology can facilitate natural language evolution within a geographically dispersed community — a model applicable to any minority language community facing vocabulary gaps in specialized domains. The community-driven approach contrasts with top-down standardization efforts and aligns with how spoken languages naturally develop technical vocabulary through use and consensus. The paper also provides practical lessons for designing video-based collaboration platforms for deaf users, including the importance of minimizing upload friction, supporting multiple participation levels, and enabling rich discussion rather than simple ratings. The finding that community-driven resources outpace formal dictionaries by years validates the approach of empowering language communities to drive their own vocabulary development.
Tags: American Sign Language · deaf education · STEM accessibility · online collaboration · video forums · language development · crowdsourcing