- From: James Nurthen <james.nurthen@oracle.com>
- Date: Fri, 28 Jun 2013 09:37:01 -0700
- To: WCAG <w3c-wai-gl@w3.org>, Protocols and Formats Working Group WG <w3c-wai-pf@w3.org>
- Message-ID: <51CDBBAD.2000901@oracle.com>
forwarding to WCAG too as I missed the WCAG list on my original reply -------- Original Message -------- Subject: Re: [html-techs-tf] June 27, 2013 meeting cancelled due to lack of quorum. No meeting July 4, due to holiday Date: Fri, 28 Jun 2013 09:35:37 -0700 From: James Nurthen <james.nurthen@oracle.com> To: w3c-wai-pf@w3.org Rich, On 6/28/2013 8:29 AM, Richard Schwerdtfeger wrote: > > Hi James, > > I tried to respond to the survey but I get this message on the survey > when I try to respond: > > "You are allowed to read the questionnaire, but you are *not > authorized to answer* with your current credentials!" > The group is set up in WCAG space so you need to be a WCAG member to answer the survey. Unfortunately this seems to be the way with joint TFs. If you wish to join WCAG we can probably arrange that. > > > Given that I cannot respond to the survey I only had an issue with the > first failure criteria. This would have been in my comments: > > "User Agents should be able to determine a link from the text of its > contents. One example shows an image with alt text on it as a child of > the link. this should be enough to determine the name of the link. > It is enough to give the link a name. The point is that the name chosen fails 2.4.4 as it does not include any context as to what you are downloading in a progamatically determinable manner. > The HTML5 should specify the name computation on its elements. This > would be a namefromcontent computation and if it is not supported in > browsers it should be. ARIA name computation will default to alt on an > image if no aria labelling metadata is provide. " > The point of the user agent notes is to state that aria-label and aria-labelledby don't seem to be very well supported by user agents. This seems to be an AT bug rather than a browser bug though as the correct things seem to be exposed to the a11y APIs. Regards, James > > > Cheers, > Rich > > > > Rich Schwerdtfeger >
Received on Friday, 28 June 2013 16:37:37 UTC