- From: John Foliot <john@foliot.ca>
- Date: Thu, 5 Jul 2012 15:37:58 -0700
- To: "'Edward O'Connor'" <eoconnor@apple.com>, <public-html@w3.org>
Edward O'Connor wrote: > > > Do you mean that WG should choose between > > "defer 194-6 to HTML.Next" and "mint transcript with media element"? > > I expect there will be 3 Change Proposals to choose between: the two > you > mention[1][2], and a third (Silvia's "Introduction of a @transcript=URL > attribute" proposal)[3]. > > Right, the [1] proposal advocates deferring the addition of a mechanism > for programmatic association of transcripts with media elements in HTML > to HTML.next. Hi Eric, Ted, While it is certainly within the right of the Working Group to consider deferring this important accessibility requirement to HTML.next, I think that you will find that the Accessibility Task Force will find a strong objection to this suggestion, and when it comes to the (evitable?) WBS Survey those strong objections will be lodged there at that time. As such Eric, while the no-change Change Proposal appears to still be in play, in practical considerations I think that the 2 Change Proposals that actually advance a solution moving forward will be the ones to investigate and discuss more fully. > > > And the transcript will be integrated with media element, right? > > Yes, the [2] and [3] proposals advocate such a feature, though with a > somewhat different design. One of the concerns that I and others had during the lengthy discussions we've had on this topic was of "integration", both in terms of how access to a transcript would be seamlessly communicated and acted upon by all users (including users with disabilities). There emerged an implied understanding that any solution should/would allow for a "discovery, selection and activation method" (a.k.a. a menu item/switch) that would be part of the media controls. The other concern was over the placement of the actual link-to-transcript, and here both proposals have different possible ways of doing this, but both allow for the linkage to be enclosed within the opening and closing <video> (and <audio>) tags: one because @transcript would take a direct URL, the other would be by allowing the <a href="transcript.html" id="foo"> link inside of the element: <video transcript="foo"> <a href="transcript.html" id="foo"> </video> We wanted to be sure that in the use-case where third party authors did a "view source, copy and paste" of a video, that all of the required assets (and associated URLs) be captured at that time. There will also need to be some good authoring guidance to go hand-in-hand with this (for example the recommendation to always specify fully resolved URLs, rather than relative paths). More specifics are in each Draft CP. > > > And how about 194-2C, what's the relationship between 2C and > > alt/longdesc of media element? > > I assume you're referring to [2]. This is a proposal for ISSUE-194, not > for ISSUE-203. Technically correct, however there has been some apparent confusion over the roles of "long textual descriptions" of a video asset versus the "transcript" of a video asset. Issue 194 is dealing with the programmatic linkage of a Transcript to a <video> element, while Issue 203 seeks to ensure that a method is present that allows for a short and long textual description of the movie itself: Short (accessible name): Oceans 11 Long (accessible description): "Danny Ocean and his eleven accomplices plan to rob three Las Vegas casinos simultaneously. Stars: George Clooney, Brad Pitt and Julia Roberts. Director: Steven Soderbergh." Transcript (alternative presentation): "One con under escort. Open gate one. Man walking. Open gate two. - Let's go. Face the wall. - Got one under escort. - Good morning. - Morning. - State your name for the record. - Daniel Ocean. - Thank you. The purpose of this hearing is to determine..." (Poster description*): "The movie's main characters (listed from foreground to background and right-to-left: George Clooney, Brad Pitt, Matt Damon, Andy Garcia, and Julia Roberts) are standing outside of what appears to be a Las Vegas casino, however the marquee reads "Ocean's Eleven". The poster also reads: "3 casinos, 11 guys, 150 million bucks. Ready to win big?" along the top, and then finally "Place your bets" at the bottom left corner." http://www.moviegoods.com/Assets/product_images/1020/189540.1020.A.jpg (* we do not yet have a means of delivering on this requirement) > > Both nonzero-edit proposals ([2] and [3]) allow for linking to HTML or > text files, regardless of what server is hosting what. > > Any method of indirection will be more complex than a direct link, but > there are several other design considerations that argue for using an > indirect link for these use cases, as I argue in [2]. > > > So I think using URL is better, and what is important difference or > > rational for 2 options of transcript? > > I hope the text of all three change proposals is clear enough to > establish the design tradeoffs of the different approaches. If it's > not, > please provide feedback so Silvia and/or I can improve our proposals! +1. Both Ted's and Silvia's proposals have strengths and weaknesses, and so collectively we need to examine and weigh up those differences and choose which we believe best meets our requirements. From an accessibility perspective, both proposals meet the use-case requirements we have brought forward from the Accessibility Task Force level. Cheers! JF
Received on Thursday, 5 July 2012 22:38:36 UTC