Idea11y: Enhancing Accessibility in Collaborative Ideation for Blind or Low Vision Screen Reader Users
Mingyi Li, Huiru Yang, Nihar Sanda, Maitraye Das · 2026 · Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems (CHI '26) · doi:10.1145/3772318.3790878
Summary
Digital whiteboards such as Miro and FigJam have become central to brainstorming, design sprints, and UX research, yet their reliance on freeform spatial layout, drag-and-drop manipulation, and rapid visual scanning leaves blind and low-vision (BLV) screen reader users largely excluded. The authors designed Idea11y, a Miro plug-in that transforms whiteboard content into a hierarchical, editable text outline (Frame → Cluster → Note) by algorithmically applying gestalt principles (proximity, similarity, enclosure) to detect implicit clusters, then presents the result in a header/sub-header/bullet structure compatible with standard screen-reader navigation (e.g., H and Shift+H in JAWS/NVDA). The system adds AI-generated cluster summaries (GPT-4o-mini), voice coding that assigns distinct synthesized voices to different creators or colors, earcons and spoken alerts for collaborator co-presence, jumping shortcuts to collaborator cursor locations, and an accessible voting mode. The paper is grounded in a formative study with eight sighted whiteboard users that identified four key strategies (externalising ideas via sticky notes, spatial clustering, shared-feedback evaluation, and cursor-based collaboration awareness), then evaluated Idea11y through individual sessions with 13 BLV screen reader users and collaborative sessions with six BLV-sighted dyads. The work is framed through Guilford's divergent/convergent thinking, the Geneplore model, Shneiderman's creativity support framework, and Bennett et al.'s interdependence framework for ability-diverse collaboration.
Key findings
Twelve of thirteen BLV participants preferred Idea11y over Miro. Compared to the Miro baseline, Idea11y produced statistically significant improvements for understanding how notes were grouped (p<0.05), understanding where a newly added note was located (p<0.01), moving notes (p<0.01), editing note text (p<0.05), and ease of learning the manipulation actions (p<0.01). The hierarchical outline made cluster structure comprehensible and reduced the repeated-reading and skipped-note problems that plagued Miro's arrow-key navigation. Voice coding was appreciated by 6 of 13 participants for differentiating creators and colors (mean ease 4.15 and 4.00 respectively), though tracking many voices alongside meeting audio introduced cognitive load. Earcons and co-presence alerts supported collaboration awareness on demand without interrupting reading flow, but push notifications were rated mixed (mean 2.92) and some participants disabled them. In the dyadic study, BLV participants added an average of 5.2 notes — sometimes more than their sighted partners (e.g., B2: 9 vs S2: 5) — and scored Idea11y higher than sighted partners scored Miro on CSI Expressiveness (7.50 vs 6.92) and Immersion (5.58 vs 5.33). Limitations surfaced: the linear outline sacrificed spatial reasoning, and sighted collaborators could not see BLV screen reader focus on the Miro canvas.
Relevance
For practitioners building or procuring collaborative tools, this paper is a concrete demonstration that accessible ideation is tractable without abandoning the creative affordances of digital whiteboards — the key move is algorithmically surfacing the implicit structure (gestalt clusters, co-presence, color semantics) that sighted users infer visually, then rendering it in conventional screen-reader navigation patterns. Product teams working on Miro, FigJam, Mural, Jamboard, or any freeform-canvas tool can adopt the specific mechanisms (hierarchical text outline linked two-way to the canvas, voice coding for metadata, earcons for co-presence, jumping shortcuts, accessible voting) as a near-term backlog. The paper is also valuable for researchers reconsidering visual-centric creativity frameworks (Geneplore, Shneiderman) and for accessibility teams supporting ability-diverse collaboration in design, education, and consulting contexts. Limitations include a small sample and English-only evaluation, and the system currently covers sticky notes only — not images, sketches, or arrows.
Tags: blind and low vision · screen readers · digital whiteboards · collaborative ideation · creativity support · ability-diverse collaboration · gestalt principles · auditory cues · voice coding · brainstorming