# Native popover enhancement with disclosure fallback

Study ID: `popover-fallback-study-2026-09-16`

## Question

In one controlled browser fixture, can a release note remain readable when a native Popover enhancement is used and when that enhancement is deliberately bypassed?

## Test fixture

The fixture starts with an open HTML `details` disclosure containing the release note. Its script checks whether the test element exposes a `showPopover()` function.

- Enhancement run: when the function is present, the fixture hides the disclosure, reveals a button, invokes the button, and checks the open popover state with `:popover-open`.
- Fallback run: a query parameter deliberately bypasses the enhancement. The original open disclosure remains in the document.

Both paths look for the same sentence: “Keyboard settings now stay with the current browser profile.” A run passes only when that task text is present and the expected presentation state is observable.

## Conditions

- Collected September 16, 2026.
- Google Chrome 153.0.8010.48 in headless mode.
- macOS 26.6.2, build 25G83.
- Viewport: 1280 by 900 CSS pixels at device scale factor 1.
- Local file fixture; no remote page or network data was used.
- Runner: `npm run study:popover-fallback` under Node 22.

The runner launches an isolated Chrome profile for each path, reads the fixture’s JSON observation through Chrome’s debugging protocol, captures the viewport, and closes the browser. The source fixture is `scripts/studies/popover-fallback-fixture.html`; the runner is `scripts/run-popover-fallback-study.mjs`.

## Recorded outputs

- `observations.json`: complete environment and boolean observations.
- `observations.csv`: one flat row per exercised path.
- `enhancement-open.png`: native popover state.
- `fallback-open.png`: disclosure fallback state.

## Limits

This is a component behavior check in one Chrome build, not a cross-browser compatibility study, accessibility audit, user study, or performance benchmark. The fallback run proves the base branch in the same modern browser. It does not emulate a browser without Popover support, a script failure, an assistive-technology combination, or keyboard focus behavior. The result applies only to this fixture and recorded environment.

## Dataset reuse

The original values and field arrangement in `observations.json` and `observations.csv` are dedicated to the public domain under [CC0 1.0](https://creativecommons.org/publicdomain/zero/1.0/). This applies only to Useful Tomorrow’s original observation data. It excludes third-party documentation and content, browser and software names, logos, trademarks, and other third-party rights.

## Version and corrections

- Version 1.0 — September 16, 2026: initial fixture, two observed paths, raw JSON and CSV, and two screenshots.
- Corrections: none as of September 16, 2026. Future substantive corrections will identify the changed claim and update the study’s modified date.