- From: Andrew Kirkpatrick <akirkpat@adobe.com>
- Date: Wed, 8 Aug 2018 19:51:25 +0000
- To: Jonathan Avila <jon.avila@levelaccess.com>, "Patrick H. Lauke" <redux@splintered.co.uk>, "w3c-wai-gl@w3.org" <w3c-wai-gl@w3.org>
- Message-ID: <9BCFF04D-A6BD-4E57-9779-B9B7A9B6092C@adobe.com>
I think that the greatest problem with relying on placeholder is with 3.3.2 - I believe that the group felt that a label needs to be persistent and that the fact that a placeholder used as a label will disappear when the field is filled can trigger a failure. Thanks, AWK Andrew Kirkpatrick Head of Accessibility Adobe akirkpat@adobe.com http://twitter.com/awkawk On 8/8/18, 15:27, "Jonathan Avila" <jon.avila@levelaccess.com> wrote: > We've identified that placeholder use as the sole form of labelling is strongly discouraged, but I believe we do at least partly agree here that the fact that it is used as a last resort for the accessible name calculation is correct? If it is a last resort for the accessible name -- then when only it is used the component still has an accessible name and assuming it's a meaningful name it passes WCAG. I don't agree with that putting in the acc name even as a last resort because it legitamizes it's use as an accessible name and makes it difficult for us to prevent it's use and proliferation. I'd prefer to see something like -- is not used in the accessible name calculation but browser's may expose it as fallback content. Jonathan -----Original Message----- From: Patrick H. Lauke [mailto:redux@splintered.co.uk] Sent: Wednesday, August 08, 2018 3:08 PM To: w3c-wai-gl@w3.org Subject: Re: Bug: Firefox Accessibility Inspector reports placeholder attribute as eligible for accessible name On 08/08/2018 19:46, Jonathan Avila wrote: >> And, in Firefox and Chrome at least (possibly others, no time to test) the placeholder IS exposed by the browser as the input's programmatically determinable / accessible name. So how is it failing? > > If the placeholder is exposed as the name but doesn't provide a name for the field but rather an example value then it would fail some SC because the programmatic name isn't a name but rather something else. In the same we would fail an input for date who's aria-label was "mm/dd/yyyy". Yes, but I was responding specifically to the example Glenda provided, where the placeholder was used with "First name" as value... Getting back to the original topic: placeholder is currently one of the last resort attributes used to provide an accessible name to a control, in the absence of anything more suitable like a <label>, an aria-label, aria-labelledby or even a title attribute. We've identified that placeholder use as the sole form of labelling is strongly discouraged, but I believe we do at least partly agree here that the fact that it is used as a last resort for the accessible name calculation is correct? 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
Received on Wednesday, 8 August 2018 19:51:59 UTC