RE: VTT edge margins: we need to know current practice. PLEASE RESPOND If you implement a VTT UA (even in a polyfill)

Hi Nigel,

CIL…

Best,
John

John Birch | Strategy and Business Development Manager | Screen
Main Line : +44 1473 831700 | Ext: 2208 | Direct Dial: +44 1473 834532
Mobile: +44 7919 558380 | Fax : +44 1473 830078
John.Birch@screensystems.tv<mailto:John.Birch@screensystems.tv>

Visit us at
BVE, Excel London, Stand N25, 27 Feb - 1 Mar

[http://www.subtitling.com]<http://www.subtitling.com/>  [https://www.linkedin.com/company/screen-subtitling-systems-ltd] <https://www.linkedin.com/company/screen-subtitling-systems-ltd>   [https://www.youtube.com/channel/Screen Subtitling Systems] <https://www.youtube.com/channel/UCBp2nyUeNbIFz9cD66Ym3AQ>   [https://twitter.com/ScreenSystems] <https://twitter.com/ScreenSystems>

[cid:BuyNowButtonBanner450x128_f51819c7-6de4-47de-8a98-b6f818ff5db0.jpg]<http://subtitling.com/products/subtitle-create/create/wincaps-q4-subtitling-software/>
P Before printing, think about the environment
From: Nigel Megitt [mailto:nigel.megitt@bbc.co.uk]
Sent: 16 November 2017 11:28
To: John Birch <John.Birch@screensystems.tv>
Cc: singer@apple.com; public-texttracks@w3.org; TTWG <public-tt@w3.org>
Subject: Re: VTT edge margins: we need to know current practice. PLEASE RESPOND If you implement a VTT UA (even in a polyfill)

John,

If I understand correctly those values are relative to the encoded video rather than the video viewport.  Is that correct?
Yes, they are expressed in video pixels in the base video resolution. I’m not quite sure what you mean by encoded video and video viewport (perhaps you are referring to overscan – which is not really relevant in broadcast anymore?) We (at Screen) do not have any practical concept of a separate canvas for subtitles… we tend to think in terms of the effective changes to the underlying video pixels. In fact this is quite a relevant point in terms of subtitle creation… the traditional practise is to lock the subtitle pixels over the obscured video pixels… because subtitlers during the creative process traditionally have a responsibility to avoid important on screen action in the video, or to avoid lower thirds, or to place subtitle text into areas of the video specifically left ‘black’ for subtitles etc. The concept of a subtitle canvas or viewport that is an independent layer in the presentation over the underlying video has introduced new concepts (and complexity) into subtitling ☺

Just to reiterate my concern here, since I'm not sure how well it came over in the minutes and issue notes: by defining an inset area relative to the video viewport only the ability to position subtitles and captions relative to the video is removed. Specifying some expected or normal values for an inset within the viewport area does not resolve this, though it could help to set some expectations that might slightly reduce the problem, on a good day with a following wind.

By the way, stating an aesthetic motivation for not allowing caption areas to sit against the viewport edge is a subjective point that should be in the realm of the caption author not the specification. There are likely to be exceptions that should not be prohibited.

I agree… and the original motivations for subtitle placement away from the boundary were NOT primarily aesthetic, they were practical. In the first instance early broadcast could not guarantee the presentation of pixels near the edges of the transmitted video signal (the overscan issue)… more recently, with the introduction of HD and the use of ‘eye tracking’ research, it has been ‘determined’ that a more centred presentation enhances the reception (the readability) of the subtitles on larger display devices – so this placement is still a practical consideration rather than an aesthetic one.

Nigel
--
Nigel Megitt
Executive Product Manager, BBC Design & Engineering
Telephone: +44 <tel:+44%C2%A03030807996> (0)3030807996<tel:+44%C2%A03030807996>
BC2 C1 Broadcast Centre, Media Village, 201 Wood Lane, London W12 7TP<x-apple-data-detectors://3>

On 14 Nov 2017, at 16:28, John Birch <John.Birch@screensystems.tv<mailto:John.Birch@screensystems.tv>> wrote:
By my calculation for SD that is ~7%... [50/720 * 100]
As previously for HD there are practical reasons (not just aesthetic reasons) for placing the margins closer to the centre of the wider picture.

I suspect there is no scope within your algorithm for a sensitivity to the rendering area aspect ratio - in which case a value of #7% would also work... but result in an atypical presentation (from a broadcasters perspective :)

Best regards,
John



John Birch | Strategy and Business Development Manager | Screen
Main Line : +44 1473 831700 | Ext: 2208 |  Direct Dial: +44 1473 834532
Mobile: +44 7919 558380 |  Fax : +44 1473 830078
John.Birch@screensystems.tv<mailto:John.Birch@screensystems.tv>

Visit us at
BVE, Excel London, Stand N25, 27 Feb - 1 Mar


PBefore printing, think about the environment

-----Original Message-----
From: singer@apple.com<mailto:singer@apple.com> [mailto:singer@apple.com]
Sent: 14 November 2017 16:08
To: John Birch <John.Birch@screensystems.tv<mailto:John.Birch@screensystems.tv>>
Cc: public-texttracks@w3.org<mailto:public-texttracks@w3.org>; TTWG <public-tt@w3.org<mailto:public-tt@w3.org>>
Subject: Re: VTT edge margins: we need to know current practice. PLEASE RESPOND If you implement a VTT UA (even in a polyfill)

Hi John

it really is the exact value used in the algorithm in 6.1 for this step that I am looking for:

"       • Let edge margin be a user-agent-defined horizontal length, expressed as a percentage of the width of the video’s rendering area, which will be used to define a margin at the left and right edges of the video into which this cue will not be placed. In situations with overscan, this margin should be sufficient to place the cue within the title-safe area. In the absence of overscan, this value should be picked for aesthetics (to avoid text being aligned precisely on the left or right edge of the video, which can be ugly)."


On Nov 14, 2017, at 5:30 , John Birch <John.Birch@screensystems.tv<mailto:John.Birch@screensystems.tv>> wrote:

Hi David,

As an information point... in our systems for the generation of subtitles in DVB bitmap format or for open (burnt in) insertion we use the following starting values for subtitle 'margins'.
N.B. These are often subsequently adjusted (tweaked) on a channel by channel basis to suit local requirements.

For SD resolution...
50 pixels left and right

For HD resolution...
240 pixels left and right

I hope this is of use...
N.B. the HD values are intended to create a centred 4:3 style presentation. Even though the screen has a wider aspect ratio in HD, subtitles still tend to be maintained in a 4:3 presentation window in the screen centre to minimise eye movements - but this is up to the broadcasters... some choose to push the left and right margins closer to the screen edges. Our system, as is typical practise in subtitling, justifies to the set margins, rather than using absolute pixel positioning for the subtitles.

Best regards,
John Birch



John Birch | Strategy and Business Development Manager | Screen Main
Line : +44 1473 831700 | Ext: 2208 |  Direct Dial: +44 1473 834532
Mobile: +44 7919 558380 |  Fax : +44 1473 830078
John.Birch@screensystems.tv<mailto:John.Birch@screensystems.tv>

Visit us at
BVE, Excel London, Stand N25, 27 Feb - 1 Mar


PBefore printing, think about the environment

-----Original Message-----
From: David Singer [mailto:singer@mac.com]
Sent: 13 November 2017 23:35
To: public-texttracks@w3.org<mailto:public-texttracks@w3.org>; TTWG <public-tt@w3.org<mailto:public-tt@w3.org>>
Subject: VTT edge margins: we need to know current practice. PLEASE
RESPOND If you implement a VTT UA (even in a polyfill)

This sentence is problematic giving each UA a free hand:
"In the absence of overscan, this value should be picked for aesthetics (to avoid text being aligned precisely on the left or right edge of the video, which can be ugly)."

Resolution at TPAC: add a default edge margin instead of the above
sentence ? (find out what the different UAs have implemented right
now)


* * * *

Please let us know what you actually implement. We would like to
settle on a recommended or even mandated value for edge margin.  If we
all actually have the same value, we can choose that and not cause
disruption, so KNOWING what EVERYONE does kinda makes sure you’re not
accidentally disadvantaged…

PLEASE tell us….

David Singer

singer@mac.com<mailto:singer@mac.com>



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

David Singer
Manager, 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



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

http://www.bbc.co.uk

This e-mail (and any attachments) is confidential and may contain personal views which are not the views of the BBC unless specifically stated.
If you have received it in error, please delete it from your system.
Do not use, copy or disclose the information in any way nor act in reliance on it and notify the sender immediately.
Please note that the BBC monitors e-mails sent or received.
Further communication will signify your consent to this.

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

Received on Thursday, 16 November 2017 12:17:24 UTC