- From: Leif Halvard Silli <xn--mlform-iua@xn--mlform-iua.no>
- Date: Wed, 22 Aug 2012 11:50:22 +0200
- To: Benjamin Hawkes-Lewis <bhawkeslewis@googlemail.com>
- Cc: Leif Halvard Silli <xn--mlform-iua@xn--mlform-iua.no>, Steve Faulkner <faulkner.steve@gmail.com>, Chaals McCathieNevile <w3b@chaals.com>, John Foliot <john@foliot.ca>, Maciej Stachowiak <mjs@apple.com>, public-html@w3.org, HTML Accessibility Task Force <public-html-a11y@w3.org>
Benjamin Hawkes-Lewis, Wed, 22 Aug 2012 00:21:42 +0100: > On Wed, Aug 22, 2012 at 12:17 AM, Leif Halvard Silli wrote: >> Heh! That was in fact cool to hear. So all NVDA is lacking is the >> announcement! Authors could in fact create polyfills for the >> announcement, e.g. via aria-label or aria-labelledby, so that user can >> use it. > > They could, but not without adversely impacting usability in UA/AT > combinations that already notify users about longdesc (e.g. Firefox + > JAWS). That is true. It would have to be short and pregnant, to not disturb too much. Thus, users would have *know* the impact of hearing e.g. "Description link." May be the page could contain a option to *add* that text to the @alt text - a scripted on/off option. Or just some information saying: "The sentence 'Description link.' in the alternative text indicates there is a longdesc link available." Sidepoint: I have previously argued in favor of <a role=presentation>. And there are in fact no difference between an img whose longdesc URL can be activated if the user by accident hits the Enter key - and a link with role=presentation. It is thus interesting if NVDA developers think it is OK to not make the user aware that the img has a longdesc link that the user could trigger by accident. PS. Come to think about it: May be NVDA halfway supports the aria-describedby longdesc link as well, then ... ? >>>> * He also says that 'If we do implement this, it will only be as a >>>> temporary solution because of the lack of support in current accessible >>>> browsers'. This is a technical argument. But it sounds like a incorrect >>>> technical argument. At least today - 2 years later - where we hear that >>>> Firefox *does* support @longdesc, in the accessibility API - which I >>>> assume that NVDA would use. >>> >>> I think by support James means "discoverable to *all* users, not just >>> users of assistive technology". So just having a hook in the >>> accessibility API doesn't count. >> >> Yes. But the world is not going to move if we are all waiting on each >> others. > > James may feel that @longdesc is not going to be worth discovering or > providing if it's not universally discoverable. Yup. I feel he is not correct there. > I doubt Firefox is waiting on NVDA here; after all JAWS was exposing > @longdesc long before NVDA. I agree - Firefox follows its own track. But if I have understood Jonas correctly previously, then the support is not new. It is just that they have 'aria-describedby-ified' their @longdesc implementation a little bit, lately. If I got it right. -- leif halvard silli
Received on Wednesday, 22 August 2012 09:51:16 UTC