WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: Misuse of TabIndex 0

for

From: Patrick H. Lauke
Date: Jul 7, 2017 1:19PM

On 07/07/2017 20:02, Sailesh Panchang wrote:
> Patrick,
> Your view now seems to differ from the one you expressed in Nov /2015
> when you seemed to agree with Michael re. 'operability' besides his
> other assertions?

You mean I may have changed my mind in the course of the last 1.5 years?
Perhaps so. My main point is: while I would generally note issues like
non-interactive elements receiving focus in 2.4.3, I wouldn't
categorically state that these are hard FAILs, exactly because the
language of WCAG doesn't explicitly say that non-interactive elements
should NEVER receive focus. This is the wooly part of WCAG where - often
based on context, personal reasoning, interpretation - we have to apply
some creative "under what WCAG SC can I fail this thing which is not
good but doesn't have a clear SC related to it".

> We are not discussing broken keyboard operability for elements that
> are indeed operable ... so I do not understand application of 2.1.1 to
> non-operable element with tabindex=0 now.

See above. As the fact that non-interactive elements fall into the
keyboard focus cycle falls across various SCs, without clearly - in my
view - being a hard FAIL, it can - depending on what other issues under
2.1.1 there may already be - be something that I'd mention there as well
as part of a more general "there are issues relating to how this
site/app operates or doesn't via the keyboard". Again, point being: it's
not a clear cut failure that falls very squarely into a neat single SC,
so sometimes you'll need to be a bit creative on where you ding a site
for doing this.

> I do not see what you state for visible focus indicator as part of the
> normative text of that SC. The Understanding doc covers 'operable'
> elements.

So what are you saying? That if the focus is on an inoperable element,
there should NOT be a visible indication of focus?

2.4.7 relates to "keyboard operable user interface". It does not say
"operable elements" specifically, nor does it explicitly say "if an
element receives focus but it is not operable, it should not have a
focus indication". Leaving aside things that shouldn't be forced into
focus with tabindex=0 that aren't interactive...what about read-only
text fields? Do you consider them "not operable", and therefore exempt
from requiring visible focus indication under 2.4.7? Presumably not, and
under that same rationale I'm saying that 2.4.7, to paraphrase, says:
"sighted user must be able to see where the current focus is". It does
not make any value judgement about what should or should not RECEIVE
focus in the first place, it merely says that focus needs to be visibly
discernible.

And to avoid doubt: I'm not saying adding non-interactive things into
the focus order with tabindex=0 is good practice, far from it. What I
*am* saying is that in my view there's not a clear-cut "well, this fails
2.4.3 / 2.4.7" situation, based on my current reading of WCAG. I usually
still fail this sort of thing in audits, but it's not necessarily a
"hard" fail per the exact letter of WCAG, more of a judgement call about
exactly where/how you note the failure.

P
--
Patrick H. Lauke

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