- From: John Foliot <john@foliot.ca>
- Date: Wed, 13 Aug 2014 07:14:01 -0700
- To: "'Adam Powell'" <adam@adaminfinitum.com>, "'Shawn Henry'" <shawn@w3.org>
- Cc: "'WAI Interest Group'" <w3c-wai-ig@w3.org>
Adam Powell wrote: > > > I'm replying to this working group, rather than the HTML > Accessibility Task Force list because I have a question > about what I should do to contribute. > > I am unfamiliar with the phrase 'Call For Implementations." > > It sounds as if we should create webpages that follow the > draft's details/specifications so that it can be tested but > I don't recall ever having seen this step before. > > I reviewed the linked pages for guidance but all the mentions > of "implementation" seem to be guidance for authors and user > agents. > > Can anyone clarify? Hi Adam, Thanks for the question. Your reading is indeed correct (although here it is somewhat skewed by the fact that we already have some implementations of @longdesc). You are correct, in that what it means is that it is an invitation to start using @longdesc: if you are a content author, and you are adding a complex image to your page (say for example a complicated graph or chart, or an infographic), then as an author, please go ahead and start using @longdesc, knowing that now it is a fully valid and compliant part of HTML5. If you are a browser vendor, assistive technology manufacturer, or other "user-agent" tool maker (this includes authoring tools too, BTW), then from that perspective please go ahead and tool/re-tool your application to support @longdesc, as once again it is on track to being a full part of HTML5 (Candidate Recommendation is the final step on the standards journey, before becoming a full Recommendation - which is W3C speak for Standard). The phrase you are tripping up on is the 'Call for Implementations' - this is standard language for announcements like this, and is normally applied when there is a "new" feature added. It signals the fact that there is already emergent support for that new feature: it is a mechanism to try and break the 'chicken and egg' problem. It is W3C's way of saying go ahead, there is enough support today to start using it safely. In the case of @longdesc, which was an existing attribute in HTML4 that was not originally being supported in HTML5, this extension spec corrects that problem, by ensuring that @longdesc remains valid when moving from HTML 4 to HTML5. Support in most of the mainstream screen readers exists today (JAWS, NVDA, WindowEyes, and a number of other non-North American screen readers, the exception being VoiceOver), along with any number of browser extensions, which extends the usefulness of @longdesc for users with other requirements (cognition issues perhaps?) - those plugins are also exposing the @longdesc content to the end-user. The single largest "problem" with @longdesc today has always been, well, actually 2 problems: the chicken and egg problem (authors would not use @longdesc, for lack of mainstream browser support; browsers would not support due to lack of usage), and poorly crafted / incorrect usage of the @longdesc attribute (coupled with the "link-rot" problem of long ago[*]). Collectively we are seeking to right those wrongs, and your contributions, as an author, user-agent implementer (or even consumer - if you consume a provided longdesc description - send an email to the author and thank them!) will help move this useful attribute forward. Hope this answers your question, feel free to ping back if anything is unclear. Cheers! JF ------------------------------ John Foliot Web Accessibility Specialist W3C Invited Expert - Accessibility Co-Founder, Open Web Camp [* see: http://john.foliot.ca/wysiwyg_longdesc/ ]
Received on Wednesday, 13 August 2014 14:14:35 UTC