-
Notifications
You must be signed in to change notification settings - Fork 125
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add Aria-repeated #1188
Add Aria-repeated #1188
Conversation
closes issue 1044
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for doing this! Some nits and a couple of wording changes for your consideration.
@@ -13471,6 +13535,7 @@ <h2>Substantive changes targetted for the 1.3 release</h2> | |||
<h2>Substantive changes since the last public working draft</h2> | |||
<ul> | |||
<!-- EdNote: After each WD publish, move contents of this list into the next one below. --> | |||
<li>03-Feb-2020: Add <pref>aria-repeated</pref>.</li> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for adding the ChangeLog entry! Based on experience, this tends to result in merge conflicts because things don't land right away and/or in the expected order. Therefore, James and I typically do the ChangeLog after. (I'm not saying don't include it; merely giving you the option.)
Looks like my other comments got lost :( I'll rewrite. |
index.html
Outdated
<div class="property-description"> | ||
<p>Indicates content is repeated for visual purposes.</p> | ||
<p>The <code>aria-repeated</code> helps indicate that content is repeated, and that presenting the repeated content to assistive technology users could produce underesirable redundancy in the output. Unlike <pref>aria-hidden</pref>, <code>aria-repeated</code> content may be presented by screen readers to users when requested by the user.</p> | ||
<p>Authors SHOULD NOT use <code>aria-repeated</code> on the very first instance of the repeated content, and SHOULD use <code>aria-repeated</code> on subsequent redundant instances of the same content. Setting <code>aria-repeated</code> on a subtree indicates the entire subtree is redundant, and it cannot be reversed within that subtree by setting <code>aria-repeated="false"</code> on a descendant.</p> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What if the very first instance of the repeated content is a pull quote which, for aesthetic reasons, appears first?
In addition, I assume that authors MAY use aria-repeated
on the very first instance of the repeated content -- IF they set it to false
.
Therefore what about something in the ballpark of:
Authors SHOULD only set <code>aria-repeated</code> to <code>true</code>
on redundant instances of content because assistive technologies MAY skip
over these instances during presentation to the user.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ok, I took this but wanted to keep the part about aria-repeated=false.
Could we actually disallow the false value completely? I don't see how it's useful.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm good with this change, but I'm not resolving the conversation yet because you ask about disallowing the false
value completely. That might be a discussion worth having with the group.
I personally think disallowing false
on a boolean would be weird. So either false
is the default value that no one uses because why bother? Or aria-repeated
shouldn't be a boolean. IMHO.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
If I have a submit button at the beginning and one at the end of a form, wouldn't it be better to use aria-repeated=true
on the first submit button?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Probably should tell authors not to apply it to elements that contain interactive content. I would not want my screen reader to hide either submit button regardless of what repeated is set to.
<pdef>aria-repeated</pdef> | ||
<div class="property-description"> | ||
<p>Indicates content is repeated for visual purposes.</p> | ||
<p>The <code>aria-repeated</code> helps indicate that content is repeated, and that presenting the repeated content to assistive technology users could produce underesirable redundancy in the output. Unlike <pref>aria-hidden</pref>, <code>aria-repeated</code> content may be presented by screen readers to users when requested by the user.</p> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Nit: "The aria-repeated property helps..." (Or delete the "The")
<p>Indicates content is repeated for visual purposes.</p> | ||
<p>The <code>aria-repeated</code> helps indicate that content is repeated, and that presenting the repeated content to assistive technology users could produce underesirable redundancy in the output. Unlike <pref>aria-hidden</pref>, <code>aria-repeated</code> content may be presented by screen readers to users when requested by the user.</p> | ||
<p>Authors SHOULD only set <code>aria-repeated</code> to <code>true</code> on redundant instances of content because assistive technologies MAY skip | ||
over these instances during presentation to the user. Setting <code>aria-repeated</code> on a subtree indicates the entire subtree is redundant. If <code>aria-repeated</code> is true for a subtree, it cannot be reversed within that subtree by setting <code>aria-repeated</code> to false on a descendant.</p> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Nit: code
tags for "true" and "false" since you're using them as values.
<li>Repeated table header rows</li> | ||
<li>Pull quote</li> | ||
</ul> | ||
<p>Because the repeated content may be important in some scenarios, e.g. during proofreading, assistive technologies MAY provide options so that screen readers users can choose whether or not to filter <code>aria-repeated</code> content out of the current view.</p> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
s/screen readers users/screen reader users/
What if area with |
<pdef>aria-repeated</pdef> | ||
<div class="property-description"> | ||
<p>Indicates content is repeated for visual purposes.</p> | ||
<p>The <code>aria-repeated</code> helps indicate that content is repeated, and that presenting the repeated content to assistive technology users could produce underesirable redundancy in the output. Unlike <pref>aria-hidden</pref>, <code>aria-repeated</code> content may be presented by screen readers to users when requested by the user.</p> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
content may be presented by screen readers
If it's a normative "may" (which is what I think you meant?) then it needs to be MAY.
If not, then please change it to something like:
content can be presented by screen readers
<li>Repeated table header rows</li> | ||
<li>Pull quote</li> | ||
</ul> | ||
<p>Because the repeated content may be important in some scenarios, e.g. during proofreading, assistive technologies MAY provide options so that screen readers users can choose whether or not to filter <code>aria-repeated</code> content out of the current view.</p> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
repeated content may be important
This instance of lowercase 'may' isn't normative, but we have recently decided to try not to use lowercase 'may', 'must' or 'should' in the prose just in case it confuses anyone. (There's usually a decent alternative word that doesn't mess up the meaning). So the above could be rewritten as either:
repeated content might be important
or:
repeated content can be important
<pdef>aria-repeated</pdef> | ||
<div class="property-description"> | ||
<p>Indicates content is repeated for visual purposes.</p> | ||
<p>The <code>aria-repeated</code> helps indicate that content is repeated, and that presenting the repeated content to assistive technology users could produce underesirable redundancy in the output. Unlike <pref>aria-hidden</pref>, <code>aria-repeated</code> content may be presented by screen readers to users when requested by the user.</p> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
<p>The <code>aria-repeated</code> helps indicate that content is repeated, and that presenting the repeated content to assistive technology users could produce underesirable redundancy in the output. Unlike <pref>aria-hidden</pref>, <code>aria-repeated</code> content may be presented by screen readers to users when requested by the user.</p> | |
<p>The <code>aria-repeated</code> helps indicate that content is repeated, and that presenting the repeated content to assistive technology users could produce undesirable redundancy in the output. Unlike <pref>aria-hidden</pref>, <code>aria-repeated</code> content may be presented by screen readers to users when requested by the user.</p> |
over these instances during presentation to the user. Setting <code>aria-repeated</code> on a subtree indicates the entire subtree is redundant. If <code>aria-repeated</code> is true for a subtree, it cannot be reversed within that subtree by setting <code>aria-repeated</code> to false on a descendant.</p> | ||
<p>Here are some examples of repeated content:</p> | ||
<ul> | ||
<li>Running headers and footer sections, repeated on each page</li> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
<li>Running headers and footer sections, repeated on each page</li> | |
<li>Running header and footer sections, repeated on each page</li> |
Should |
I have quite a few editorial comments, but it is hard to figure out what they ought to be with all the unresolved suggestions. So, I'll wait on the word-smithing. Other concerns:
|
Good point @mcking65. Ignore my previous question about whether complementary role should support aria-repeated - it shouldn't. If a user doesn't want to hear repeated content, then they don't want it to be a landmark, either. I guess the DPUB-ARIA doc-pullquote example should probably use a div instead of an aside. Hmmm, that example currently uses aria-hidden to hide the repeated aside from AT. I wonder what it means to use aria-hidden on a landmark? They seem to be in conflict. Do AT respect the landmark, or the aria-hidden? |
@carmacleod wrote:
Nothing overrides aria-hidden. However, there are the unspecified author error recovery behaviors where focusability overrides naming, not readability, in some browsers that we discussed today. |
I wonder if we have the right name on this feature. Is this really limited to completely redundant content? The use cases all seem related to enabling some kind of reader mode. The objective seems to be to indicate that certain content should be ignored when in that kind of mode. What if it were ...
|
Maybe we need something more like |
I want to wait on resolving suggestions until we all talk about it during our weekly meeting. |
This is interesting |
This might be my favorite. |
I think we want to be a little bit careful about giving authors the ability to tell ATs to skip content. I think ARIA should be more descriptive rather than prescriptive. Let's provide information about the content and give the ATs good information to improve user experience. |
I like that sentiment. I'm not sure if requiring us to be able to point to the id of the redundant content will really help:
At least with this proposed property (whatever we call it), the screen reader will be able to provide the user an option to read it anyway. |
@aleventhal is this still needed? If not please close it. |
I don't need it now that we have doc-pageheader, doc-pagefooter and doc-pagenum. |
closes #1044
Preview | Diff