W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > July to September 2001

RE: Caption synchronization tolerance

From: Charles McCathieNevile <charles@w3.org>
Date: Tue, 28 Aug 2001 21:58:05 -0400 (EDT)
To: "Charles F. Munat" <chas@munat.com>
cc: "GLWAI Guidelines WG (GL - WAI Guidelines WG)" <w3c-wai-gl@w3.org>, Larry Goldberg <larry_goldberg@wgbh.org>, <geoff_freed@wgbh.org>
Message-ID: <Pine.LNX.4.30.0108281433310.22369-100000@tux.w3.org>
Chas,

there are many use cases where the captions aren't the same length as the
content, or where it is desirable to pause the rest of the presentation -
this is also the case in audio-described content, where it takes extra time
to put in the audio descriptions.

SMIL 2 was designed to handle these cases better in various ways. For
example, it is possible to have audio descriptions (for example) which, if
they are played, force the presentation to wait at key points (this is
related to the way that it controls synchronisation failure, but also to
some more purpose-designed functions it introduces).

Of course there are other cases where this is not desirable. Enabling a user
to choose between slowing the overall presentation, and keeping it running in
real time, is a nice choice to have when people are getting a live
transmission. The technology exists (not just in SMIL, but also with
conventional TV) for making this choice at multiple points, and going from
one mode  to the  other by various means (fast-forward through buffered
content, or just jump to current real-time point, or ...)

cheers

Charles McCN

On Mon, 27 Aug 2001, Charles F. Munat wrote:

  Why do captioned objects need to be the same length as uncaptioned objects?
  Why wouldn't it be possible (and even desirable) to allow captioned versions
  to pause the audiovisual long enough for caption readers to catch up?

  Just wondering.

  Chas. Munat
Received on Tuesday, 28 August 2001 21:58:19 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:12 GMT