Re: ISSUE-446: Should be a subset of IMSC1 [TTML WebVTT mapping note]

Hi Pierre,

As background, here is what we were thinking when we defined the profile in the mapping document.  The initial idea behind defining a profile in the mapping document was to use it as a way of structuring the mapping process.  To that end, the T2VP ( now changed to TVTT in the doc, in response to Issue-415 ) profile was defined in relation to the WebVTT functional set, not IMSC or other TTML profiles.   Having the profile breaks the mapping process up into two phases:

1.  General TTML1.0 -> T2VP ( now TVTT ) Profile
2.  T2VP  ( now TVTT ) -> WebVTT

 This two step approach has a couple of advantages:

1.  Any documents produced at the interim stages of the mapping process conform to TTML.  There’s never a point where you have a document that doesn’t conform to either TTML or WebVTT.  This allows for incremental testing along the way, among other benefits.

2.  The second step is really just about doing the semantic translation from TTML to WebVTT.  All the unit conversions and structural transformations or reorganizations were handled in the first step.

Our goal was to provide a strategy for mapping any existing TTML document to WebVTT.  It’s interesting to see your perspective on it, where you are thinking about newly authored content, and the benefits of having a profile that could be used to author content in TTML that would easily map to WebVTT.

When Andreas is back from vacation, I’m sure he will want to contribute to this discussion.

Courtney

> On Sep 23, 2015, at 3:55 PM, Pierre-Anthony Lemieux <pal@sandflow.com> wrote:
> 
> Hi David and Nigel,
> 
> For one thing, as mentioned in the ticket, selecting an existing
> profile of TTML as a starting point simplifies the mapping task, e.g.
> no need to consider #markerMode or #pixekAspectRatio. In addition, in
> the case of IMSC1, it also allows (limited) support of CSS features
> like background-image, not available in vanilla TTML1.
> 
> Best,
> 
> -- Pierre
> 
> On Wed, Sep 23, 2015 at 3:49 PM, David Singer <singer@apple.com> wrote:
>> 
>>> On Sep 23, 2015, at 4:10 , Nigel Megitt <nigel.megitt@bbc.co.uk> wrote:
>>> 
>>> I disagree that the T2VP profile should be deliberately constrained to be
>>> a subset of IMSC1, because it would be too limiting and wasn't specified
>>> in the charter description of this Note.
>> 
>> I’m with Nigel.  Well, mostly. No, no, I support him.
>> 
>> Let’s work out the best mapping, and then if the mapped part of TTML is not cleanly part of a current profile, think about how to handle that case. So I am sort of with both of you: I don’t want to constrain the mapping work a priori, but I also don’t want to emerge with a needless mismatch a posteriori.
>> 
>>> 
>>> I do agree that if there are significant things that prevent IMSC 1 from
>>> being mappable to WebVTT then we should look at the discrepancies and try
>>> to reconcile them. I also would not object to an additional set of
>>> constraints that explain what limitations would apply to WebVTT in order
>>> to convert it to IMSC 1, extra to what could be done if mapping to TTML1
>>> more generally.
>>> 
>>> Some obvious things that aren't in WebVTT but are constraints of IMSC 1
>>> that I do not think we need to carry over are the reference fonts, the
>>> simultaneously active region limit and the HRM.
>> 
>> Yup.
>> 
>>> 
>>> Nigel
>>> 
>>> 
>>> On 23/09/2015 05:51, "Timed Text Working Group Issue Tracker"
>>> <sysbot+tracker@w3.org> wrote:
>>> 
>>>> ISSUE-446: Should be a subset of IMSC1 [TTML WebVTT mapping note]
>>>> 
>>>> http://www.w3.org/AudioVideo/TT/tracker/issues/446
>>>> 
>>>> Raised by: Pierre-Anthony Lemieux
>>>> On product: TTML WebVTT mapping note
>>>> 
>>>> The profile/authoring guidelines specified in the document should be a
>>>> subset of IMSC1:
>>>> 
>>>> - it helps narrow down the set of features to be supported, e.g. no need
>>>> to support #markerMode
>>>> - it simplifies the job of implementers who are already supporting IMSC1
>>>> 
>>>> Features have been deemed necessary but are not in IMSC1 should be
>>>> considered for addition there.
>>>> 
>>>> 
>>>> 
>>> 
>>> 
>>> 
>>> -----------------------------
>>> 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.
>>> -----------------------------
>> 
>> David Singer
>> Manager, Software Standards, Apple Inc.
>> 
>> 
>

Received on Wednesday, 23 September 2015 23:54:02 UTC