Consider removing role=group guidance in complex images
shawna-slh opened this issue · 5 comments
from e-mail:
https://www.w3.org/WAI/tutorials/images/complex/
I would avoid role="group" in your code examples. It overwrites the semantics of the figure element. Testing with jaws 2019, there is no sematic announced at all in your example.
I have tested the examples with Chrome 79 and JAWS 2019: the role of the group is output correctly when reading with the virtual cursor, and the role of the graphic is always output. In this respect, I do not think using role=group
is problematic, especially since <figure>
is not always output correctly by JAWS
Group
is an allowed value, but I think nowadays we could probably default to the figure
role and not overwrite the default. That should give AT users a more consistent experience. (The figure
role did not exist when we wrote the tutorials in 2014 – it is new in ARIA 1.1 that was released in 2017 and we have not significantly updated our guidance since then. We’re currently in the process of moving the content into the new design and making updates.)
Allowed ARIA role attribute values:
figure
role (default - do not set),group
orpresentation
.
https://www.w3.org/TR/html52/ & https://w3c.github.io/html-aria/#el-figure
I have added an editor’s note to the editor’s draft: https://wai-tutorial-images.netlify.com/tutorials/images/complex/#approach-3-structurally-associating-the-image-and-its-adjacent-long-description-html5
Also, duplicate of: https://github.com/w3c/wai-tutorial-images/issues/2
Just discovered this issue as I was trying to understand the discrepancy between the WAI tutorial's advice vs. Scott O'Hara's recommendation from https://www.scottohara.me/blog/2019/01/21/how-do-you-figure.html - adding the link here, as I found it to be a valuable reference documenting some tests results.
...is the suggested repetition of the figcaption
's text in aria-label
something to include in this tutorial as well?