W3C home > Mailing lists > Public > public-tt@w3.org > January 2012

RE: TTML after March

From: John Birch <john.birch@screen.subtitling.com>
Date: Mon, 16 Jan 2012 11:26:14 -0000
Message-ID: <4476B296B92A4741A49B0BD01759070001D35283@sss-uk-ex-02.screen.subtitling.local>
To: "Glenn Adams" <glenn@skynav.com>, "public-tt" <public-tt@w3.org>
Cc: "David Singer" <singer@apple.com>
Hi Glenn,


Thanks for your clarification of the role of XSL-FO as a descriptive mechanism for TTML.

This has particular relevance to a recent discussion in EBU-DFXP about centre left/right justification… (a combined justification method used in certain subtitling applications).

It seems difficult to me, how you can express this justification concept using XSL-FO as it would seem to require an ’inside out’ calculation of extents. But I must confess I have only a cursory understanding of XSL-FO.


As you clarify, new formatting attributes in TTML (or a derivative!) are clearly not constrained to using XSL (or CSS) models for their definition, although clearly being able to express such an attribute in terms of the XSL-FO model is desirable.


On the subject of TTML WG resurrection, the issues that currently are most of interest to me are:


The role of the XSD in TTML definition – I know the EBU-DFXP group have had some problems with this.

The profile feature

Extension to TTML through the metadata element


Best regards,



From: Glenn Adams [mailto:glenn@skynav.com] 
Sent: 16 January 2012 07:14
To: public-tt
Cc: David Singer
Subject: Re: TTML after March


As David alludes, the whole business of TTML dependency on XSL-FO is based on a poor reading of the spec. The only use made (and intended to be made) of XSL-FO was for the purpose of defining formatting semantics without having to do so directly in TTML. Further, for the most part, the style properties adopted in TTML are properties which XSL-FO borrows from CSS. In doing this, TTML does not make use of the CSS Box Model, but instead makes use of the XSL-FO Area Model, which are distinct, but similar models (and conceptual vocabularies) for describing formatting and presentation.


It is true, however, that TTML explicitly avoided the use of CSS grammars (except for property value grammars for those properties adopted from CSS via XSL-FO). The rationale for this choice was to avoid having to have two different parsers: an XML parser and a CSS parser.


If there were sufficient industry interest in doing so, and, in particular, if this would produce a positive impact on TTML take-up, then it would be relatively straightforward to define a CSS binding for TTML that employs CSS for style association rather than the existing mechanism defined in TTML.



On Mon, Jan 16, 2012 at 12:04 AM, Glenn Adams <gadams@xfsi.com> wrote:


John Birch | Screen Subtitling Systems Ltd | Strategic Partnerships Manager
Main Line : +44 (0)1473 831700 | Ext : 270 | Office :  
Mobile: +44 (0)7919 558380 | Fax: +44 (0)1473 830078
john.birch@screen.subtitling.com | www.screen.subtitling.com | 
The Old Rectory, Claydon Church Lane, Claydon,Ipswich,IP6 0EQ,United Kingdom

Visit us at
BVE, Stand K56, Earls Court, London, 14-16 Feb 2012

Before Printing, think about the environment

---------- Forwarded message ----------
From: David Singer <singer@apple.com>
Date: Wed, Jan 11, 2012 at 5:37 PM
Subject: Re: TTML after March
To: Philippe Le Hegaret <plh@w3.org>
Cc: public-tt <public-tt@w3.org>

I'd like to move to a CG for a while.  I think that there is some work to be done; the one that comes immediately to mind is documenting the profile of TTML as she is used.  A full-on WG may not be needed.

On Jan 11, 2012, at 13:40 , Philippe Le Hegaret wrote:

> The charter of the TTML Working Group will expire at the end of March
> 2012 [1] (it was extended for one year in 2011). I'm curious about the
> thoughts from individuals here about what to do, if anything, beyond the
> end of March.
> For example, SMPTE did some extensions to SMPTE-TT back in 2010. Should
> we look at those and fold them back in the specification?


> Dynamic flow was removed from the specification due to lack of
> implementation experience. Should it be reconsidered?
> Should we switch from XSL FO to CSS?

or document how CSS styling can be used.  People think that XSL is intrinsic, and I am not sure the group agrees.

> Should we do a profile of TTML as well and retaining features that are
> the most deployed?

Yes!  This is urgent!

> Or should the group just declare victory and go home?
> Thank you,
> Philippe
> [1] http://www.w3.org/2008/01/timed-text-wg.html
> [2]
> http://www.w3.org/TR/2010/CR-ttaf1-dfxp-20100223/#style-attribute-dynamicFlow
> PS: if someone wants to ask the Timed Text Community Group for feedback
> on this matter as well, feel free to forward this message. Input is
> welcome!

David Singer
Multimedia and Software Standards, Apple Inc.



This message may contain confidential and/or privileged information. If you are not the intended recipient you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation. Screen Subtitling Systems Ltd. Registered in England No. 2596832. Registered Office: The Old Rectory, Claydon Church Lane, Claydon, Ipswich, Suffolk, IP6 0EQ
Received on Monday, 16 January 2012 11:27:22 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 5 October 2017 18:24:05 UTC