Keyboard accessibility
Ability to operate functionality without requiring a pointing device. Successful tabbing through a page does not prove logical order, visible focus, or complete task access.
Glossary · area 19 of 19
What a page owes a keyboard, a screen reader and a fact-checker: the accessibility basics and the C2PA vocabulary for proving where a file came from.
22 terms. Each opens its own page with what it can and cannot support, how the scanner measures it, and where it comes up in the guides.
Ability to operate functionality without requiring a pointing device. Successful tabbing through a page does not prove logical order, visible focus, or complete task access.
The sequence in which interactive elements receive keyboard focus. It should preserve meaning and operability rather than merely matching DOM appearance.
A visible cue showing which component currently receives keyboard input. Presence alone is insufficient if contrast, size, or occlusion makes it imperceptible.
Accessible Rich Internet Applications attributes and roles that communicate interface semantics to assistive technologies. Incorrect ARIA can make native semantics worse, and it does not change visual behavior by itself.
Additional text conveying a control’s purpose or detail beyond its accessible name. It should supplement rather than duplicate or replace the name.
Difference in relative luminance between foreground and background. Passing one sampled pair does not establish compliance across states, overlays, gradients, and images.
A user preference requesting less nonessential animation and movement. Respecting it reduces motion-triggered barriers but does not fix flashing or inaccessible controls.
Synchronized text alternatives for speech and relevant non-speech audio in media. Automatically generated captions require accuracy review; presence alone is not adequacy.
Narration conveying important visual information not available from the primary audio. It addresses visual content, not captions for spoken content.
A change in content communicating results, progress, errors, or state without necessarily moving focus. Assistive technology needs programmatic exposure to announce it appropriately.
The preferred nontechnical term for a C2PA Manifest containing signed provenance assertions bound to an asset. It can show tamper-evident history but not whether the content is truthful.
A structured statement about an asset included in a C2PA Manifest. A signature protects the assertion from undetected alteration; it does not independently verify its claim.
The signed manifest component that binds assertion references and related information. Valid structure and signature establish integrity under the trust model, not factual accuracy.
The digital signature protecting a C2PA claim. It verifies key-based authorship and integrity when trust validation succeeds, not copyright ownership.
The container holding one or more C2PA Manifests associated with an asset. A file can lose embedded manifests through platform processing or metadata stripping.
Information cryptographically associating digital content with a specific C2PA Manifest. It detects mismatched content but cannot prevent copying or editing.
A cryptographic binding derived directly from asset bytes or structure. Material modification can invalidate it even when the edit is legitimate.
A robust identifier, fingerprint, or watermark supporting manifest rediscovery after transformations. Its resilience introduces matching thresholds and possible false associations.
A prior asset incorporated into a new C2PA asset and referenced through provenance assertions. An ingredient relationship records reuse without settling whether the reuse was authorized.
A maintained collection of accepted credential issuers or signing authorities used during verification. Inclusion supports signer trust decisions; it does not endorse every assertion that signer makes.
A region of a page with a declared role: header, nav, main, footer, aside. Screen readers use landmarks to jump between regions, so a page with no main landmark forces a reader through everything. They are structural, cost nothing, and are absent on a surprising share of pages that pass a contrast check.
The luminance ratio between text and its background, which WCAG requires to be at least 4.5 to 1 for normal text and 3 to 1 for large text. It must hold in hover and focus states as well as at rest, which is where most passing pages fail; a check that reads only the resting colours misses it.