E-mail List Archives

Search E-mail List Archives for

Number of posts in this thread: 18 (In chronological order)

From: Vaibhav Saraf
Date: Mon, Oct 05 2020 4:33AM
Subject: Bypass blocks for a small website
No previous message | Next message →

Hi Everyone,

We are working on a small website having 5 main pages. When a user is on
any page on the website the focus is managed to take him/her directly to
the main section of the page.

If the user moves to the top of the webpage, he/she will need to go through
7 tab strokes (2 in header & 5 in navigation bar) to reach the main
section.

In such a case does it become mandatory to provide a bypass mechanism to
the main content? How will it change if the on page load user is at the top
of the page and not in the main content?

Thanks,
Vaibhav

From: Patrick H. Lauke
Date: Mon, Oct 05 2020 4:53AM
Subject: Re: Bypass blocks for a small website
← Previous message | Next message →

On 05/10/2020 11:33, Vaibhav Saraf wrote:
> Hi Everyone,
>
> We are working on a small website having 5 main pages. When a user is on
> any page on the website the focus is managed to take him/her directly to
> the main section of the page.
>
> If the user moves to the top of the webpage, he/she will need to go through
> 7 tab strokes (2 in header & 5 in navigation bar) to reach the main
> section.
>
> In such a case does it become mandatory to provide a bypass mechanism to
> the main content? How will it change if the on page load user is at the top
> of the page and not in the main content?

I would actually say that it's not really ideal to auto-focus/yank the
focus to the main part of the page on load. Users may not be aware/know
that this happened, and may completely miss the content that came before
the bit where you forced their focus to be on load. So I'd recommend
removing this. And then yes, add a simple skip link to allow user to
jump directly to the main content.

P
--
Patrick H. Lauke

https://www.splintered.co.uk/ | https://github.com/patrickhlauke
https://flickr.com/photos/redux/ | https://www.deviantart.com/redux
twitter: @patrick_h_lauke | skype: patrick_h_lauke

From: Vaibhav Saraf
Date: Mon, Oct 05 2020 5:17AM
Subject: Re: Bypass blocks for a small website
← Previous message | Next message →

Hi Patrick,

Thanks for the suggestion. I understand your concern regarding the
auto-focus in this case and will surely take it up in my team.

Thanks,
Vaibhav


On Mon, 5 Oct 2020 at 16:23, Patrick H. Lauke < = EMAIL ADDRESS REMOVED = >
wrote:

> On 05/10/2020 11:33, Vaibhav Saraf wrote:
> > Hi Everyone,
> >
> > We are working on a small website having 5 main pages. When a user is on
> > any page on the website the focus is managed to take him/her directly to
> > the main section of the page.
> >
> > If the user moves to the top of the webpage, he/she will need to go
> through
> > 7 tab strokes (2 in header & 5 in navigation bar) to reach the main
> > section.
> >
> > In such a case does it become mandatory to provide a bypass mechanism to
> > the main content? How will it change if the on page load user is at the
> top
> > of the page and not in the main content?
>
> I would actually say that it's not really ideal to auto-focus/yank the
> focus to the main part of the page on load. Users may not be aware/know
> that this happened, and may completely miss the content that came before
> the bit where you forced their focus to be on load. So I'd recommend
> removing this. And then yes, add a simple skip link to allow user to
> jump directly to the main content.
>
> P
> --
> Patrick H. Lauke
>
> https://www.splintered.co.uk/ | https://github.com/patrickhlauke
> https://flickr.com/photos/redux/ | https://www.deviantart.com/redux
> twitter: @patrick_h_lauke | skype: patrick_h_lauke
> > > > >

From: wolfgang.berndorfer
Date: Mon, Oct 05 2020 8:19AM
Subject: Re: Bypass blocks for a small website
← Previous message | Next message →

And additionally to the skip link make sure that the main section is really
a <main> section and starts with an H1. These features also support
efficient navigation with screen readers. And they are a small effort on a
small page.

From: Anna Lewis
Date: Mon, Oct 05 2020 8:34AM
Subject: Re: Bypass blocks for a small website
← Previous message | Next message →

Hello,

I'm not sure how I got added to this list. Can you please remove me?

Thank you,
Anna

On Mon, Oct 5, 2020 at 8:19 AM < = EMAIL ADDRESS REMOVED = > wrote:

> And additionally to the skip link make sure that the main section is really
> a <main> section and starts with an H1. These features also support
> efficient navigation with screen readers. And they are a small effort on a
> small page.
>
>
>

From: Birkir R. Gunnarsson
Date: Mon, Oct 05 2020 9:34AM
Subject: Re: Bypass blocks for a small website
← Previous message | Next message →

WCAG 2.4.1 only requires mechanisms to bypass block of content (one or
more of a skip link, headings and ARIA landmarks).
So technically if the page has one heading or landmark it would pass
(in the case of one landmark it would be borderline pass since ARIA
requires that all content on the page resides in a landmark region).
In your case, yes, I would use landmark regions for head/main/footer
and leave focus at the top of the page when it loads (the only
possible expceiotn being if there is a login page you may want to
autofocus the username or the first input).
The skip link is probably a good idea. For me, I use 10 tab stops as a
rule of thumb indicator how important it is, but I generally recommend
a global skip link on websites, even with as few as 5 tab stops.



On 10/5/20, Anna Lewis < = EMAIL ADDRESS REMOVED = > wrote:
> Hello,
>
> I'm not sure how I got added to this list. Can you please remove me?
>
> Thank you,
> Anna
>
> On Mon, Oct 5, 2020 at 8:19 AM < = EMAIL ADDRESS REMOVED = > wrote:
>
>> And additionally to the skip link make sure that the main section is
>> really
>> a <main> section and starts with an H1. These features also support
>> efficient navigation with screen readers. And they are a small effort on
>> a
>> small page.
>>
>>
>>

From: glen walker
Date: Mon, Oct 05 2020 10:54AM
Subject: Re: Bypass blocks for a small website
← Previous message | Next message →

I was going to say (almost) the same thing as Birkir. 2.4.1 just says a
"mechanism" is available to bypass blocks. If your mechanism is to
autofocus to main, that technically passes.

Whether that's a good idea from a UX perspective is a different discussion.

However, regarding landmarks, I'm a bit OCD when it comes to normative vs
non-normative. Birkir said ARIA *required* landmarks. Where is that
stated? We all know the first rule of ARIA is to not use ARIA (
https://www.w3.org/TR/aria-in-html/#rule1) and that ARIA "is intended to
provide missing semantics so that the intent of the author may be conveyed
to assistive technologies" (https://www.w3.org/TR/wai-aria-1.1/#ua-support).
If we focus on using native HTML instead of ARIA, then it seems like the
<main> element would be a required element, but it's not. I get a little
nervous when we use the term "required" unless I can find the reference in
a normative doc.

Feel free to slap me down, Birkir, if I missed something regarding the
requirement of landmarks. That's what's great about this WebAIM forum, I'm
always learning something new no matter how long I've been doing it.

From: Patrick H. Lauke
Date: Mon, Oct 05 2020 11:40AM
Subject: Re: Bypass blocks for a small website
← Previous message | Next message →

On 05/10/2020 16:34, Birkir R. Gunnarsson wrote:
> WCAG 2.4.1 only requires mechanisms to bypass block of content (one or
> more of a skip link, headings and ARIA landmarks).

In practice though, headings and ARIA landmarks are only of use to AT
users, not regular keyboard users. So, pragmatically, skip links are
still fairly de-rigeur.

> (the only
> possible expceiotn being if there is a login page you may want to
> autofocus the username or the first input)

But even there you run the risk of possibly confusing AT users, who may
not immediately realise that there's content/stuff before that
autofocused form field.

P
--
Patrick H. Lauke

https://www.splintered.co.uk/ | https://github.com/patrickhlauke
https://flickr.com/photos/redux/ | https://www.deviantart.com/redux
twitter: @patrick_h_lauke | skype: patrick_h_lauke

From: wolfgang.berndorfer
Date: Tue, Oct 06 2020 4:28AM
Subject: Re: Bypass blocks for a small website
← Previous message | Next message →

A plead for usability awareness:

I couldn't find out, what OCD stands for, but I know the abbreviation UX.

2.4.1 has purely usability in mind. 11 keyboard strokes don't make a webpage
inaccessible stricte dictum. They make it just annoying.

So let's not only reflect the "skip links" to satisfy the required effort to
pass 2.4.1. Let's point out, what can be done to improve UX for AT: Headings
and regions.

"Although this Success Criterion deals with blocks of content that are
repeated on multiple pages, we also strongly promote structural markup on
individual pages as per Success Criteria 1.3.1. "
(https://www.w3.org/TR/UNDERSTANDING-WCAG20/navigation-mechanisms-skip.html)

Besides: An accordion mechanism would fit too, to pass SC 2.4.1. Correct?

Wolfgang

From: Mallory
Date: Tue, Oct 06 2020 10:29AM
Subject: Re: Bypass blocks for a small website
← Previous message | Next message →

Let me fix that please:
"Let's point out, what can be done to improve UX for SCREEN READERS: Headings
and regions."

not everyone has admin privileges to install rando extensions/plugins that take things like headings and landmark regions and lets keyboarders or switch control users or whoever to ALSO use those to move focus.

On Tue, Oct 6, 2020, at 12:28 PM, = EMAIL ADDRESS REMOVED = wrote:
> A plead for usability awareness:
>
> I couldn't find out, what OCD stands for, but I know the abbreviation UX.
>
> 2.4.1 has purely usability in mind. 11 keyboard strokes don't make a webpage
> inaccessible stricte dictum. They make it just annoying.
>
> So let's not only reflect the "skip links" to satisfy the required effort to
> pass 2.4.1. Let's point out, what can be done to improve UX for AT: Headings
> and regions.
>
> "Although this Success Criterion deals with blocks of content that are
> repeated on multiple pages, we also strongly promote structural markup on
> individual pages as per Success Criteria 1.3.1. "
> (https://www.w3.org/TR/UNDERSTANDING-WCAG20/navigation-mechanisms-skip.html)
>
> Besides: An accordion mechanism would fit too, to pass SC 2.4.1. Correct?
>
> Wolfgang
>
>
>

From: glen walker
Date: Tue, Oct 06 2020 10:45AM
Subject: Re: Bypass blocks for a small website
← Previous message | Next message →

Sorry, Wolfgang. OCD is obsessive compulsive disorder. It could be
considered a "hidden" disability and like many disabilities, can have a
wide range or spectrum of intenseness.

I was using the term in referring to myself in that when talking about
what's "required" by accessibility, I get very nit-picky about normative vs
non-normative statements. I was not short-coming best practices but I
think it's important to distinguish between what's really required from a
normative perspective vs what's a best practice from a non-normative
perspective. Sometimes legal implications can ride on the difference. I
always encourage best practices and hopefully companies are focusing on the
full UX and not just a checklist of what's required legally.

So wrapping this back to Birkir's comment about landmarks being required, I
have not found anything normative that says this. Are landmarks helpful?
Absolutely. Should every page have at least a main landmark, and probably
a header and footer landmark? Absolutely. Is it required? No, at least
not with my understanding of the normative requirements. I'm certainly
open to changing my mind if I've missed something in the reams of
documentation out there.

From: Vaibhav Saraf
Date: Tue, Oct 06 2020 11:45AM
Subject: Re: Bypass blocks for a small website
← Previous message | Next message →

Hi Birker,

A screen reader user will have multiple options given you have the heading
navigation, landmark navigation, tab and normal arrow key navigation on
such a page.

Consider the case for a keyboard user who lands in the middle of the page
all because auto-focus have to do a non-intuitive action of 7 Shift tabs to
log out of the site. What will be the usability impact in such a case? Do
shortcuts like Ctrl + Home work for a normal keyboard user as well?

Thanks,
Vaibhav





On Mon, 5 Oct 2020 at 21:04, Birkir R. Gunnarsson <
= EMAIL ADDRESS REMOVED = > wrote:

> WCAG 2.4.1 only requires mechanisms to bypass block of content (one or
> more of a skip link, headings and ARIA landmarks).
> So technically if the page has one heading or landmark it would pass
> (in the case of one landmark it would be borderline pass since ARIA
> requires that all content on the page resides in a landmark region).
> In your case, yes, I would use landmark regions for head/main/footer
> and leave focus at the top of the page when it loads (the only
> possible expceiotn being if there is a login page you may want to
> autofocus the username or the first input).
> The skip link is probably a good idea. For me, I use 10 tab stops as a
> rule of thumb indicator how important it is, but I generally recommend
> a global skip link on websites, even with as few as 5 tab stops.
>
>
>
> On 10/5/20, Anna Lewis < = EMAIL ADDRESS REMOVED = > wrote:
> > Hello,
> >
> > I'm not sure how I got added to this list. Can you please remove me?
> >
> > Thank you,
> > Anna
> >
> > On Mon, Oct 5, 2020 at 8:19 AM < = EMAIL ADDRESS REMOVED = >
> wrote:
> >
> >> And additionally to the skip link make sure that the main section is
> >> really
> >> a <main> section and starts with an H1. These features also support
> >> efficient navigation with screen readers. And they are a small effort on
> >> a
> >> small page.
> >>
> >>
> >>

From: wolfgang.berndorfer
Date: Tue, Oct 06 2020 11:59AM
Subject: Re: Bypass blocks for a small website
← Previous message | Next message →

Thanks Glen for information and clarification.

I agree that a distinction between normative and moral requirements is
necessary. But I'd like to add the requirement of an interpretation of
normative materia, especially when the publishing date is sensitive for the
content.

So what would the authors of SC 1.3.1 have fixed as required, if landmarks
were available in 2008?

On the other hand: Why is there still no explicit requirement? (I didn't
follow the discussions during the development of WCAG 2.1 and 2.2 very
well.)

But probably the WAI list and GitHub would be the right place for such a
discussion.

Wolfgang

From: glen walker
Date: Tue, Oct 06 2020 12:10PM
Subject: Re: Bypass blocks for a small website
← Previous message | Next message →

From a normative perspective, using auto-focus satisfies 2.4.1. It is a
mechanism for bypassing the repeated content. It works for keyboard users.

If you're asking, "how does a keyboard user get back to the bypassed
stuff?", they must use shift+tab. Ctrl+home just scrolls the screen to the
top and does not move the keyboard focus. (Actually, home/end scroll the
screen. If you use ctrl plus home/end, it does the same thing. Ctrl does
not affect the scrolling of the page.)

As far as what is the usability impact of having to do that, that's more of
a usability test. If you only have a handful of tabs (Birkir's rule of
thumb was 10 or less), then that might not be a burden, but that's a
subjective call. The fewer tab stops the better for some users. Other
users can press and hold the tab (or shift+tab) to navigate through a lot
of items.

On Tue, Oct 6, 2020 at 11:45 AM Vaibhav Saraf < = EMAIL ADDRESS REMOVED = >
wrote:

> Hi Birker,
>
> A screen reader user will have multiple options given you have the heading
> navigation, landmark navigation, tab and normal arrow key navigation on
> such a page.
>
> Consider the case for a keyboard user who lands in the middle of the page
> all because auto-focus have to do a non-intuitive action of 7 Shift tabs to
> log out of the site. What will be the usability impact in such a case? Do
> shortcuts like Ctrl + Home work for a normal keyboard user as well?
>
> Thanks,
> Vaibhav
>
>
>

From: Vaibhav Saraf
Date: Tue, Oct 06 2020 12:18PM
Subject: Re: Bypass blocks for a small website
← Previous message | Next message →

Thanks Glen.


On Tue, 6 Oct 2020 at 23:41, glen walker < = EMAIL ADDRESS REMOVED = > wrote:

> From a normative perspective, using auto-focus satisfies 2.4.1. It is a
> mechanism for bypassing the repeated content. It works for keyboard users.
>
> If you're asking, "how does a keyboard user get back to the bypassed
> stuff?", they must use shift+tab. Ctrl+home just scrolls the screen to the
> top and does not move the keyboard focus. (Actually, home/end scroll the
> screen. If you use ctrl plus home/end, it does the same thing. Ctrl does
> not affect the scrolling of the page.)
>
> As far as what is the usability impact of having to do that, that's more of
> a usability test. If you only have a handful of tabs (Birkir's rule of
> thumb was 10 or less), then that might not be a burden, but that's a
> subjective call. The fewer tab stops the better for some users. Other
> users can press and hold the tab (or shift+tab) to navigate through a lot
> of items.
>
> On Tue, Oct 6, 2020 at 11:45 AM Vaibhav Saraf < = EMAIL ADDRESS REMOVED = >
> wrote:
>
> > Hi Birker,
> >
> > A screen reader user will have multiple options given you have the
> heading
> > navigation, landmark navigation, tab and normal arrow key navigation on
> > such a page.
> >
> > Consider the case for a keyboard user who lands in the middle of the page
> > all because auto-focus have to do a non-intuitive action of 7 Shift tabs
> to
> > log out of the site. What will be the usability impact in such a case? Do
> > shortcuts like Ctrl + Home work for a normal keyboard user as well?
> >
> > Thanks,
> > Vaibhav
> >
> >
> >
> > > > >

From: glen walker
Date: Tue, Oct 06 2020 12:21PM
Subject: Re: Bypass blocks for a small website
← Previous message | Next message →

Wolfgang, keep in mind that WCAG is technology agnostic. It's not going to
specifically point out how to implement something because that would be
dependent on the technology.

You can't "interpret" normative, or rather, you can't add details to the
normative doc that isn't there. In your example for 1.3.1, all 1.3.1 says
is " Information, structure, and relationships conveyed through
presentation can be programmatically determined or are available in text."

Does that mean landmarks are required? I think it's dangerous to assume
that. Is there a "relationship" implied in the visual aspect of the page
that indicates that part of the page should be grouped together, for
example with a landmark? Perhaps. That's where the interpretation would
come in. You have to decide/interpret if there's a relationship (or
structure or information) implied. If so, then that relationship needs to
be programmatically determinable. One way to make that relationship
programmatically determinable is to use a landmark. But is it required?
Not really. This is similar to the 2.4.1 bypass blocks discussion. Skip
links are a great implementation for 2.4.1 but are they required? No.
Only a "mechanism" is required. Same with 1.3.1. The relationship needs
to be programmatically determinable. That doesn't mean the only way to do
that is with a landmark.

On Tue, Oct 6, 2020 at 12:00 PM < = EMAIL ADDRESS REMOVED = > wrote:

> Thanks Glen for information and clarification.
>
> I agree that a distinction between normative and moral requirements is
> necessary. But I'd like to add the requirement of an interpretation of
> normative materia, especially when the publishing date is sensitive for the
> content.
>
> So what would the authors of SC 1.3.1 have fixed as required, if landmarks
> were available in 2008?
>
> On the other hand: Why is there still no explicit requirement? (I didn't
> follow the discussions during the development of WCAG 2.1 and 2.2 very
> well.)
>
> But probably the WAI list and GitHub would be the right place for such a
> discussion.
>
> Wolfgang
>
>

From: wolfgang.berndorfer
Date: Wed, Oct 07 2020 10:07AM
Subject: Re: Bypass blocks for a small website
← Previous message | Next message →

Glen, you caught me: I think too much in html. But that's, what works best
for SR, and I'm a SR user.

From: Guy Hickling
Date: Sun, Oct 11 2020 2:44PM
Subject: Re: Bypass blocks for a small website
← Previous message | No next message

This thread has brought up one or two other topics as well. So I have just
started a new thread for one of them, to take it further while keeping this
one simple. See "Does SC1.3.1 require landmarks?". I would welcome your
thoughts on the new thread.
Regards, Guy