Re: Media TextFormats Accessibility Comparison Summary Released

On Fri, 17 Dec 2010 19:15:27 +0100, Janina Sajka <>  

> Colleagues:
> The HTML-A11Y Task Force Media Subteam has been examining the
> suitability, as of this writing,  of TTML and WebSRT for satisfying the  
> accessibility support
> described in our User Requirements document at
>; and
> refined during TPAC 2010 in
> and following.
> We summarize our comparison by denoting a list of strengths and a list
> of remaining issues requiring resolution in order to meet accessibility
> requirements. The comparison summary document is available at:
> We forward our analysis as input to remaining work in HTML 5
> specifications, noting that further work may, in fact, be assigned to a
> new working group in W3C.

Some comments on two of the issues mentioned with WebSRT and how they are  
are said to be solved for TTML:

1. "Work required to support hierarchical structure navigation not just  
through JavaScript."


The TTML box says that for <track kind="chapter"> a TTML file can be  
interpreted as a hierarchical set of navigation points. The suggested  
markup is:

   <body  role="x-nav-work">
     <div role="x-nav-section" >
       <p role="x-nav-section">
         <span role="x-nav-section" dur="11.300s">Index point 1.1.1 </span>
         <span role="x-nav-section" dur="20.100s">Index point 1.1.2 </span>
         <span role="x-nav-section" dur="12.900s">Index point 1.1.3 </span>
         <span role="x-nav-section" dur="13.700s">Index point 1.1.4 </span>

(dur="..." is probably supposed to be begin="...")

2. "Work required to support a means to tell the player to stop the video  
and wait for the end of speech synthesis."

which seems to refer to  

The suggested TTML solution is <p begin="1s" dur="1t"  
role="x-extended-description>This description pauses the video"</p>

Ignoring the syntax errors, the key ingredient is  

Both suggested solution for TTML are not actually defined by TTML, rather  
they are suggestions on how TTML could be extended with various roles.  
None of the x- roles are part of current the TTML spec.

If we are comparing how the formats could be extended to support various  
use cases, here are both issues solved for WebSRT:

Nested chapters:

00:00.000 --> 10:00.000
Section 1
<00:00.000>Section 1.1
<05:00.000>Section 1.2

10:00.000 --> 20:00.000
Section 2
<10:00.000>Section 2.1
<15:00.000>Section 2.2

(This gets you 1 level of nesting, if we really need more some other  
syntax is needed, but syntax is the easy part.)

Pause for speech synth:

00:01.000 --> 00:02.000 EXTENDED-DESCRIPTION
This description pauses the video at 00:02.000 if speech synth isn't done  
by then.

Summary: Suggested extensions for TTML or WebSRT shouldn't be represented  
as support in the format. If extensions are needed, WebSRT has the benefit  
that it can be put in the spec itself. For TTML it's too late, so it would  
have to be another spec on top of TTML.

Philip Jägenstedt
Core Developer
Opera Software

Received on Saturday, 18 December 2010 11:29:20 UTC