contextual role concept proposal #484
Conversation
closes #375 The purpose of this proposal is both to resolve the issue that "scoped" was an undefined term that was used to indicate when certain elements would expose particular ARIA roles or not. However, in the year of noodling on this, other activities to revise what "owned" means in ARIA, the idea of a minimum role, as well as the fact there were other elements in HTML that have "scoped" or "contextual roles" that are based on heuristics, rather than defined standards, became relevant. A change to hopefully help simplify the implementations of contextual roles is that this proposal is aiming to tie the role of an element based on its a11y tree ancestry, rather than based on specific HTML elements in the DOM. This PR represents the initial draft of the proposal. Other work to explicitly call out the elements that have contextual roles, what roles they need to adjust to, and other wording revisions to replace "scoped" with the new term would still need to be made.
still working on the exact wording. also changing some placeholder short code so as to hopefully not get more build errors for presently undefined terms
index.html
Outdated
| For instance, an element that is an <STRONG>accessibility child</STRONG> of its required <STRONG>accessibility parent</STRONG> would expose a computed role that was | ||
| appropriate for its related context. Where the same element with a contextual role might be used within another markup pattern and have a different accessibility parent, | ||
| and would therefore expose a computed role appropriate for that context. |
There was a problem hiding this comment.
Word-smithing plus a few simple inline examples.
| For instance, an element that is an <STRONG>accessibility child</STRONG> of its required <STRONG>accessibility parent</STRONG> would expose a computed role that was | |
| appropriate for its related context. Where the same element with a contextual role might be used within another markup pattern and have a different accessibility parent, | |
| and would therefore expose a computed role appropriate for that context. | |
| For instance, an element that is an <STRONG>accessibility child</STRONG> of its required <STRONG>accessibility parent</STRONG> would expose a computed role for its given context, such as a list item (<code>li</code>) within an unordered list (<code>ul</code>). However, the same element might be used within a different markup pattern and could therefore expose different contextually-dependent computed roles, such as a labeled <code>section</code> element (role <code>region</code>) versus an unlabeled <code>section</code> element (role <code>generic</code>). |
Can we think of a specific example that works for both? If so that'd be better than the sectioning content examples, which are less easy to understand and care about than a simple list.
There was a problem hiding this comment.
@cookiecrook I like these suggestions so i'm going to largely pull them in - but I'm going to be changing the use of "accessible label" back to "name".
There was a problem hiding this comment.
@cookiecrook re: other examples.
What do you think about mentioning tbody here, per the whole rowgroup conversation? That'd be an example of a contextual role where it can be ignored until it is given a name or focusable.
Another example we could use in lieu of / in combination with the list item example is the summary of a details, where only under specific conditions is the summary an html-summary, and otherwise it would need to be a role=generic
There was a problem hiding this comment.
🤷 I could go either way. I don't think most authors give much thought to <tbody> etc, so I'm not sure it'd be super helpful as an example, but I also don't think it'd be harmful.
Co-authored-by: James Craig <cookiecrook@users.noreply.github.com>
Co-authored-by: James Craig <cookiecrook@users.noreply.github.com>
use computed role in both sentences, rather than switching between terms
Co-authored-by: James Craig <cookiecrook@users.noreply.github.com>
index.html
Outdated
| @@ -210,6 +210,71 @@ <h3>Exposing HTML Features That Do Not Directly Map to Accessibility APIs</h3> | |||
| <abbr title="application programming interface">API</abbr> properties</a> in the [[core-aam-1.2]]. | |||
| </p> | |||
| </section> | |||
| <section id="mapping_contextual"> | |||
| <h3>Exposing HTML Features With Contextual Roles</h3> | |||
There was a problem hiding this comment.
This heading may not accurately capture the intent of the section; particularly, the words "HTML Features" used to describe contextual roles and how they're treated. I don't know the best way to describe the new section but if you agree, some alternatives for your consideration: Contextual Roles, Treatment of Contextual Roles, Understanding Contextual Roles, Understanding Contextual Roles and Accessibility Ancestry
|
Closing, this PR has been migrated to the ARIA monorepo: |
closes #375
The purpose of this proposal is both to resolve the issue that "scoped" was an undefined term that was used to indicate when certain elements would expose particular ARIA roles or not. However, in the year of noodling on this, other activities to revise what "owned" means in ARIA, the idea of a minimum role, as well as the fact there were other elements in HTML that have "scoped" or "contextual roles" that are based on heuristics, rather than defined standards, became relevant.
A change to hopefully help simplify the implementations of contextual roles is that this proposal is aiming to tie the role of an element based on its a11y tree ancestry, rather than based on specific HTML elements in the DOM.
This PR represents the initial draft of the proposal. Other work to explicitly call out the elements that have contextual roles, what roles they need to adjust to, and other wording revisions to replace "scoped" with the new term would still need to be made.
Implementation
Preview | Diff