Colquhoun.AI
← All guides

Research Guides

Check What Covers a Focused Research Control

· Colquhoun.shop editorial

A research page can have a logical sequence of links and still leave a keyboard user wondering where they went. The missing piece may be a layer of interface sitting over the active control, not a broken link or an incorrect tab order.

This worksheet investigates that specific problem in an authorized interface or prototype. It is not a claim that Colquhoun.shop has been certified accessible, and it does not replace a complete accessibility assessment.

1. Record a journey before changing the layout

Choose a short task that crosses different parts of the page: open the source list, inspect a citation, move to a comparison and reach the download control. Write down the starting state and the controls you expect to encounter. Use harmless sample records where activating a control could change data.

Record browser, viewport dimensions, zoom and whether a notification or panel is open. A screenshot of the initial page alone is insufficient evidence for a problem that appears halfway through a journey.

2. Separate the focused control from the covering layer

W3C's Focus Not Obscured (Minimum) guidance says a component receiving keyboard focus must not be completely hidden by author-created content. It distinguishes the component from an external focus indicator and includes specific notes for user-movable and user-opened content. W3C: Understanding Focus Not Obscured (Minimum)

When a step appears to lose focus, identify both elements in the finding: the control that received focus and the interface layer covering it. Do not write only “keyboard broken.” Check whether the component is wholly hidden, partly visible or visible with an unclear focus indicator. Those observations should not be collapsed into a single pass label.

3. Make the failure reproducible

In an invented research table, a fixed summary bar sits at the bottom of the window. Moving forward from the final source link sends focus to an export button beneath that bar. The button still works, but the person testing cannot see which action is ready.

A useful issue record names the preceding link, the hidden button, the summary bar, the viewport and the exact key sequence. Include a screenshot at the failing step rather than only after scrolling manually to find the button. Label the example as a fixture, not a live product measurement.

4. Compare the same route in different states

Repeat the route with the relevant banner or panel present and absent. Traverse backward as well as forward: reaching a control from below can reveal a different overlap. Review narrow layouts separately rather than assuming that a desktop result covers a wrapped toolbar.

For user-opened content, W3C notes that revealing the focused component without advancing keyboard focus can satisfy the stated condition. Record the actual mechanism and whether it works; do not assume any dismiss button automatically resolves the issue.

5. Retest the task, not just the screenshot

After an implementation change, repeat the original route and settings before widening the test. Verify that the expected control remains identifiable and that activating it still performs the intended action. Check nearby controls for a newly introduced overlap.

Close with a specific outcome such as “the export button stayed visible throughout the recorded route.” Avoid turning one corrected overlap into a claim that every keyboard interaction, assistive technology or page state now passes.

This guide describes the built-in Web Reader and Keyword Density behavior verified for this publication. Available data depends on the page response; these tools do not measure visits or search rankings.