W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > July to September 2014

lc-2954

From: Adam Solomon <adam.solomon2@gmail.com>
Date: Tue, 30 Sep 2014 18:46:54 +0200
Message-ID: <CALKv3=hJ5vhHr7=Dy2EpDAKarzJcLO=T04m0RJsTAkP9bW2pRA@mail.gmail.com>
To: WCAG <w3c-wai-gl@w3.org>
Cc: James Nurthen <james.nurthen@oracle.com>, Joshue O Connor <joshue.oconnor@cfit.ie>
Points raised in our discussion at the wcag group meeting on 300914, to be
considered for the next group meeting:

-title attribute (and aria-label) could be used to succeed use cases where
visible labels exists (i.e. set of phone number fields - see F82, or for
that matter any time there is a visible label w/o programmatic
determination, title could be used as a replacement and that might be
considered programmatic determination because it provides the same info as
the label). So, there is good reason to leave "title" and "aria-label" in
this failure.

-there is no failure for 4.1.2 in a case where there is no visible label
and no programmatic label, i.e. in a case where there is a control with no
label whatsoever (as can happen with a select where the items in the select
list give the user an idea what to choose from). If the group feels that
this is in fact a failure, then would it be possible to include such a
failure of 4.1.2 in F68 (in which case we would need to split the test
procedure between visible labels and cases where there is no label at all,
and instances where only 4.1.2 is violated to instances where 1.3.1 is also
violated). Or, is this use case not a failure at all, since all users are
at the same disadvantage when no visible label is provided.
Received on Tuesday, 30 September 2014 16:47:22 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 16 January 2018 15:34:16 UTC