Comments on: A review of the W3C Timed Text Authoring Format

By: silvia

silvia — Fri, 26 Mar 2010 20:20:39 +0000

@JeroenW The thing is: it’s not a caption standard alone, it’s a timed text standard and timed text can be a lot more than just subtitles and captions. For example, why should we have to invent another format to do the timed annotations that YouTube does? There is not a lot of functionality in the timed annotations that TTML (as it seems to be called now) does not provide – apart from hyperlinks. Do we really need a different format for each timed text type application?

By: JeroenW

JeroenW — Fri, 26 Mar 2010 12:38:53 +0000

I actually think the decision to leave hyperlinking out of the timed text format was a good one. In the one hand it does limit the number of use cases, on the other hand it does reflect the intend of the TT language:

* Captions are a visual translation of a clip’s audio stream. When adding hyperlinks to it, the captions actually start offering additional features. This might get confusing.
* Obviously, captions will be widely used for accessibility purposes. Assistive devices will have a hard time offering an interface to interact with in-caption hyperlinks.
* The immediate use case I see for links inside captions will be advertising (like Youtube’s adsense overlays). This might stretch Timed Text from a captions into an advertising standard …

Overall, great piece. Thanks for the exhaustive overview!

By: silvia

silvia — Sun, 28 Jun 2009 22:27:22 +0000

Daniel,

Thanks for the link to SMIL. I now understand where the extra “dur” attribute comes from – because SMIL allows for example to have repeats on an event, “dur” is a delimiting attribute that provides an upper limit for repeats. Since there are no dynamics in DFXP (thank god!), “dur” is probably surplus and should be let go of.

By: Daniel

Daniel — Sun, 28 Jun 2009 21:05:43 +0000

I agree about the “end” and “dur” ambiguity. The problem stems from the fact that TimedText is making an explicit reference to the timing semantics of SMIL 2.1, but overrides the attribute grammar with a format that is very limited (i.e. no synchronization arcs, events, or access keys), without detailing the concrete semantic implications.

In SMIL, an author can use “end” and “dur” at the *same time* for an actual purpose (i.e. this is a “feature”): this mechanism offers some interesting timing effects due to precedence order of “end” over “dur” resolved values.

In other words, “end” and “dur” are redundant in TimedText, in the sense that they are bound by simple arithmetic (end = begin + dur), which is not the case in SMIL because of more complex timing semantics and the way time values are resolved (e.g. non-deterministic events, animation sandwich model).

(just from the top of my mind on Sunday evening…I hope I’m being clear enough )

http://www.w3.org/TR/2009/WD-ttaf1-dfxp-20090602/#timing-value-timeExpression

http://www.w3.org/TR/SMIL2/smil-timing.html#Timing-TimingAttributeGrammars