ARGIRA v1.4.2: Three-Layer Uncertainty Communication Prototype for Automatic Alt-Text — Accessibility & Localization Release

ARGIRA v1.4.2 is an accessibility-focused patch release of the ARGIRA prototype for automatic alt-text generation and three-layer uncertainty communication.

It is a new version of ARGIRA v1.4.1 (DOI: 10.5281/zenodo.21850529).

Accessibility-only navigation

This release refines the accessibility-only mode to reduce unnecessary screen-reader navigation and keep the essential result path concise.

In accessibility-only mode, the following non-essential elements are hidden from the accessibility navigation path:

Warning and sample-image notes. "More details" and "More information" controls and their associated content.

The essential result path remains available: image type, confidence, generated alt-text, copy, and previous/next result navigation.

Manual developer verification with Android TalkBack confirmed the revised navigation behavior in both English and Spanish variants.

This also corrects a gap in the previous accessibility-only implementation: the CSS class hiding the details panel did not extend to the button that opens it, nor to the separate "More information" button, so both remained reachable by screen readers even though their panels were hidden.

Export controls

The six bulk export actions (Copy all as text/CSV/JSON and Download TXT/CSV/JSON) are now grouped under a single native expandable "Export options" control.

When collapsed, the six individual export actions no longer add separate stops to screen-reader navigation.

The control remains available in both normal and accessibility-only modes.

The underlying export functions are unchanged.

Interface and content refinement

The release corrects the bundled sample-image count from four to three and shortens the accessibility-only mode description to reduce redundant screen-reader output.

The changes were applied consistently to the English and Spanish artifacts.

Accessibility verification carried forward from v1.4.1

Manual developer verification with Android TalkBack was performed on the forced experimental variants used in Phase 3C.

The previously documented unresolved interaction remains: the "More information" section should open through activation rather than merely receiving navigation focus.

Apple VoiceOver was not evaluated because an available Apple/VoiceOver testing environment was not available. This is an untested scope, not a failed validation.

No external user study or participant evaluation was conducted for this release.

Spanish localization carried forward from v1.4.0

The release retains two independently executable prototype artifacts:

argira_demo_v2.html — English prototype. argira_demo_es.html — Spanish-localized variant.

The variants use separate HTML artifacts rather than an in-application language selector.

Both preserve the same experimental session mechanism, identifiers, classification logic, calibration logic, thresholds, and data flow.

Structural parity between the two artifacts is preserved.

Continuity

ARGIRA v1.4.2 does not introduce a new image-classification algorithm, calibration model, or experimental condition.

It remains an experimental research artifact and is not presented as a finished product or as validated accessibility technology.

This release is intended to support continued research, reproducibility, accessibility-oriented inspection, bilingual use through separate artifacts, and future evaluation with external users and additional assistive-technology environments.

Package contents

prototype/argira_demo_v2.html — English prototype artifact. prototype/argira_demo_es.html — Spanish-localized artifact. docs/accessibility_verification_v1.4.md — accessibility verification log and VoiceOver scope note. README.md — release overview and status. CHANGELOG.md — full version history. CITATION.cff — machine-readable citation metadata. LICENSE.txt — CC BY-NC-SA 4.0 license text.