RE: ACTION-329: Revisit issue-339 to investigate potential resolution options

Dear all,

I have followed up ACTION-329. Please find here the result as it was agreed during the EBU Subtitling Group's meeting last Friday.

Best regards,

Frans


-- Response on Issue-339

1. Intention of EBU-TT-D usage of tts:overflow.

Presentation of text is the primary goal of the EBU-TT-D use cases; whilst unexpected overflow of text is undesirable, in a chain that can not be fully constrained in regards to fonts and text layout it is considered unavoidable. It is considered a worse option to clip overflowed text and potentially change the editorial meaning as read by the user than to permit presentation engines to try to display the text.

2. Interpretation of TTML1 tts:overflow wording.

EBU interpreted the text of TTML 1 SE tts:overflow as follows: 
• The region size is unaffected by overflow.
• Wrapping of text, when wrapOption=“wrap” is bounded by the size of the region.
• Line areas flowed into a region may be affected by overflow, and they may be rendered outside the region if tts:overflow=“visible”.
• We expect presentation processors to clip all content outside the region area if tts:overflow=“hidden”.
• The default behaviour if tts:overflow is unspecified is to process the document according to the initial value “hidden”, which could in some cases, that we are unable to control, result in some text being hidden.

3. Interpretation of IMSC 1 current wording.

EBU interprets the prohibition of tts:overflow from IMSC 1 documents as a directive that the initial value of tts:overflow, “hidden”, applies and that processors are therefore expected to clip text that happens to fall outside the region.

4. Mitigation approaches.

EBU would support a recommendation that documents should be authored assuming the ‘worst case scenario’, i.e. with strict clipping of content that falls out of region areas, regardless of the value of tts:overflow in the document. 

EBU proposes a change to IMSC 1 to make clear that presentation processors shall be permitted to display text that overflows the region, indicated within documents by use of tts:overflow=“visible”:

 “NOTE: Presentation processors are permitted to display text that overflows the region if tts:overflow=“visible”. “

In order to permit those processors to show overflowed text, the value “visible” must be permitted within IMSC 1 documents. This change requires that the #overflow feature be changed to MAY be permitted in IMSC 1, as requested in Issue-339.

As a minimum, even in the absence of any functionality to display or indeed clip overflowed text, presentation processors should be permitted to process documents that have a tts:overflow attribute set either to “hidden” or “visible”.

EBU would support an erratum to TTML1SE that clarifies that the intended meaning of the tts:overflow wording is as stated in 2. above. Specifically it must be clear that, for the purposes of drawing region backgrounds and calculating wrapping, the region size is unaffected by the value of tts:overflow.

4. Impact on TTML2.

EBU does not consider tts:overflow a complete solution to the rendering and presentation problems that may occur, and 
may propose, or would support in principle, overflow avoidance strategies, for example copy-fit (aka shrink to fit) algorithms, or attribute-controlled direction for processors to adjust region origin and size to maintain the presentation of text within a region.

-----Original Message-----
From: Timed Text Working Group Issue Tracker [mailto:sysbot+tracker@w3.org] 
Sent: mercredi 17 septembre 2014 14:43
To: public-tt@w3.org
Subject: ACTION-329: Revisit issue-339 to investigate potential resolution options

ACTION-329: Revisit issue-339 to investigate potential resolution options

http://www.w3.org/AudioVideo/TT/tracker/actions/329


Assigned to: Frans de Jong








------------------------------------------------------------------------------

**************************************************
This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed.
If you have received this email in error, please notify the system manager. This footnote also confirms that this email message has been swept by the mailgateway
**************************************************

Received on Tuesday, 23 September 2014 10:32:41 UTC