VGAR Page Requirements

The general rule for supporting screen readers is that all features and functionality of the site or web application MUST be supported with at least one screen reader. In order to fully support screen readers, all aspects of the page that are available to sighted users must be available to screen reader users. 

When a user accesses a page with a screen reader, the user typically examines the content for the page title, landmark regions, headings, links, and iframes if present. When a screen reader user is looking for specific elements a list of the desired elements is opened. For instance, to a find a specific heading, a list of headings is opened and the user selects the desired heading from the list.

• JAWS with Internet Explorer/Chrome
• NVDA with Firefox/Chrome
• VoiceOver with Safari
• Windows Narrator on Microsoft Edge

When the user takes an action resulting in a visual change to the screen above the focus (e.g., user entered a Visa card number and the Visa logo above the input highlights onblur), each supported screen reader MUST inform the user of this change.

The page title MUST be properly read aloud by each supported screen reader on page load or when the URI changes on Single Page Architecture web applications.

When the list of iframes is requested by the user (or the user attempts to cycle through each), each supported screen reader MUST accurately identify all Frames designed for user interaction by reading their titles.

When the list of iframes is requested by the user (or the user attempts to cycle through each), each supported screen reader MUST NOT list any iframes designed for “system use”.

When the list of Landmarks is requested by the user (or the user attempts to cycle through each), each supported screen reader MUST accurately identify all Landmark regions by role.

When the list of Landmarks is requested by the user (or the user attempts to cycle through each), and there are multiple landmarks of the same type on the screen, each supported screen reader MUST accurately identify each uniquely by reading both the role and label.

When the list of headings is requested by the user (or the user attempts to cycle through each), the headings MUST be read accurately by each supported screen reader.

Run through all use cases with each supported screen reader, confirming that all features and functionality are fully available to non-sighted users.

Confirm that when alerts are shown on screen (i.e., success, error, or status/informative) they are also read aloud by each supported screen reader.

When the user takes an action resulting in a visual change to the screen above the focus (e.g., user entered a Visa card number and the Visa logo above the input highlights onblur), each supported screen reader announces the change to the user.

Confirm that the "page" title is read aloud by each supported screen reader on page load or when the URI changes on Single Page Architecture web applications.

Bring up the list of iframes (or cycle through each if a list is not available) with each supported screen reader and confirm that all iframes designed for user interaction have their titles read aloud.

Bring up the list of iframes (or cycle through each if a list is not available) with each supported screen reader and confirm that iframes designed for "system use" are not read aloud.

Bring up the list of Landmark roles (or cycle through each if a list is not available) with each supported screen reader and confirm that all Landmark regions are accurately identified.

If there are multiple landmarks of the same type on the screen, cycle through each with each supported screen reader and confirm that each is uniquely identified by the screen reader reading both the role and label.

Bring up the list of headings (or cycle through each if a list is not available) with each supported screen reader and confirm that all headings are accurately read.