Proposed program
Six months, hardest problem first.
Nobody has publicly driven the Unreal editor by voice or by gaze. That is the unknown the whole plan turns on, so we test it in month one. Finding out in month five would be too late to do anything about it. The lowest-bandwidth inputs come last, and disabled creators start testing halfway through, not at the end.
This program is scoped and costed but not yet funded. UEMCP, the foundation it builds on, is already released.
-
Prove feasibility fast
Audit UEMCP’s tool coverage against the target editor workflows, then prototype voice-driven camera framing through it. Nobody has publicly driven the Unreal editor by voice or gaze, so this is the real unknown. We test it first and fail fast if it does not hold.
-
Extend to cinematics, ship the command grammar
New UEMCP tools for Sequencer and for richer viewport and camera framing, plus version one of the open voice command grammar: panel navigation, tool and mode selection, and exact numeric transform entry.
-
The pointing layer, and the first major user test
Eye gaze for coarse jumps, head tracking to refine, and a discrete trigger to confirm, with enlarged hit targets and snapping. First structured session with multiple disabled creators, while the design can still change.
-
Harden, and release in public
Iterate the pointing layer on test feedback, then make the mid-project open-source release of the editor-accessibility extensions and the command pack. The community gets working code halfway through, not at the end.
-
The lowest-bandwidth inputs, and onboarding
Switch and sip-and-puff scanning with a switch-scanning command palette, plus an opinionated installer. The stack today needs programmer-level setup, which is the biggest adoption risk, so lowering that barrier is a deliverable, not an afterthought.
-
Integrate, document, release
One coherent accessible-editor workflow across every modality, and an end-to-end demo: a disabled creator opens the editor, builds a scene, frames a cinematic shot in the viewport and in Sequencer, and exports a result, hands-free. Full open-source release with documentation and tutorials.
What exists at the end, and who owns it
Everything below lands under a permissive license, publicly, and in two releases, not one. The community gets working code halfway through. The wider Unreal community inherits a more accessible engine, whether or not you ever touch a wheelchair.
Editor-accessibility extensions to UEMCP
New tools for Sequencer, viewport and camera framing, released into the existing MIT-licensed repository, not a new one.
An open voice command grammar
A forkable open command pack for the Unreal Editor: panels, tools, modes and exact numeric entry.
The pointing and scanning layer
Gaze plus head tracking plus discrete confirmation, and a switch-scanning command palette for the lowest-bandwidth inputs.
An installer, docs and tutorials
So adoption does not require programmer-level setup, which is the difference between a demo and a tool people use.
Help us build it, or be one of the first to use it
OPAL is being designed and tested with disabled creators from the start, not validated at the end. If you work in Unreal with alternative input, or you want to, we want to hear how you work.