How do we want to make decisions about the acceptability of screen reader output discoverable? #647
Labels
Agenda+App Development
For discussion during the next teleconference related to development of the ARIA-AT App
documentation
Related to documentation about the ARIA-AT project or its deliverables
process
Related to processes for governing and managing the ARIA-AT project
question
Further information is requested
In issue #638 , we decided:
The test report is the "official record" that that NVDA output of "list" appropriately conveys the role listbox. We use the word convey in assertions intentionally to help people understand that the screen reader doesn't necessarily have to parrot the technical names of roles and properties. They can do translation and simplification as long as they consistently convey the meaning.
Do we need to make it easy to review and share specific decisions about the acceptability of such translation? If so, how might we do this? Maintain a dictionary of them? Make the issues where such decisions are made easily findable?
Another example ... -when does not saying "selected" convey the state selected?
The text was updated successfully, but these errors were encountered: