W3C

Timed Text Working Group Teleconference

06 Nov 2014

See also: IRC log

Attendees

Present
jdsmith, pal, courtney, Frans, nigel, Thierry
Regrets
glenn, andreas, mdolan, cyril
Chair
nigel
Scribe
nigel

Contents


<trackbot> Date: 06 November 2014

This meeting

<scribe> scribeNick: nigel

IMSC 1 review comments

https://www.w3.org/2006/02/lc-comments-tracker/34314/WD-ttml-imsc1-20140930/

pal: I've proposed a resolution on all the comments, so I can walk through them and point out the document edits
... LC-2983
... No actionable comments. Thanking DVB for input would be appropriate.

<scribe> ACTION: pal Thank DVB for input re LC-2983 [recorded in http://www.w3.org/2014/11/06-tt-minutes.html#action01]

<trackbot> Created ACTION-348 - Thank dvb for input re lc-2983 [on Pierre-Anthony Lemieux - due 2014-11-13].

pal: LC-2982
... I've added Example 4 as a code example for Forced Content.

<pal> http://www.w3.org/AudioVideo/TT/Guide2LCtracker/overview.html

pal: I propose moving LC-2982 to pending for review by the group in the next week, then moving to resolved prior to sending out.

tmichel: Yes, once we have the resolution from the WG then I or Nigel can send the response back to the commenter.

pal: In this case we could move directly to Resolved-yes, but maybe giving the group a week to resolve it and move to Pending in the meantime.

nigel: Yes please, move it to Pending.

pal: Doing that.
... LC-2975
... I thought I'd replaced all of 'subtitle document' with the TTML1 defined term 'Document Instance' - it turns out maybe not.
... I've added the defined term Document Instance too.
... I've found the two instances, fixing right now.

<scribe> ... done.

pal: That'll be corrected next time I commit to trunk.

nigel: I think that'll fix it.

pal: LC-2976
... I've revised section 5.1 to add "The method by which this association is made is left to each application."

group: seems happy with that.

pal: Moving that one to Pending.
... LC-2968 - we covered this at TPAC already and there's already a proposed resolution.
... I added an informative note on section 8.3 explaining how it can be done in TTML1 using multiple regions.
... LC-2973
... There were really two comments here. Firstly, trying to clarify what a conforming processor should do with foreign namespaces attributes and elements.
... Secondly that the specification did not forbid the introduction of new elements and attributes in the IMSC namespace outside W3C.
... In §6.3 I've added a paragraph to replicate TTML1 to state that the namespace is mutable and reserved by W3C for future expansion.
... I also added in §6.2 the same prose as in TTML1 to say that a document may contain elements and attributes that are neither
... specifically permitted nor forbidden by a profile

jdsmith: So the implication is you would tolerate extra attributes?

pal: That's right - the intent is to replicate what was in TTML1. Andreas's concern was that 'ignore' for a transformation
... processor isn't clear - does it mean 'remove' or 'keep and do nothing'?

nigel: It's possibly not clear what a processor should do if it encounters a non-IMSC element or attribute.
... Perhaps the language from TTML1 §3.2.1 Generic Processor Performance should be brought in (3rd list item).

pal: That's already brought in by reference.
... I think Andreas's issue was the term 'ignore' which might be considered to force processors to drop such entities.

jdsmith: This seems subtle but complete when you consider the TTML1 references.

pal: Marking as pending.

nigel: Andreas has an opportunity to verify this.

pal: LC-2977 Forced Display.
... The first part of the comment is that not all combinations of values are described. I've added a sentence to the second
... paragraph to cover other combinations.
... The second part of the comment is about improving the note by pointing out it only applies to non-transparent background.
... My take on this is that the note doesn't only apply in that case. Unless we're absolutely certain then I'm concerned about
... making it too narrow.

nigel: I think possibly padding could apply, but as written the comment is accurate I believe. It's still reasonable to make
... no change given how specific that circumstance is, and to respond to the commenter that this is our decision.

pal: I've slightly changed the wording of the note in the editor's draft to make it a bit easier to read.
... The biggest edit is a reference back to TTML1 to describe the circumstances in which this applies.

nigel: We could respond that 'While technically the note may only apply when a non-transparent background is applied to a region the intent has a broader applicability, to warn authors...'

pal: I'm updating it now.
... Also marking as pending.
... LC-2978 altText
... I added a document convention section to reference the conventions used in TTML1 for specifying styling attributes and XML elements.
... That should satisfy the commenter.

nigel: There's a reader-friendliness question whether it's okay to refer to another document just for document conventions, but normatively this does address the comment.

pal: So much of this document is dependent on TTML1 that the reader needs to be aware of TTML1 anyway.
... Marking as pending.
... LC-2979
... Constraints on the #timing feature. The text previously said "The same syntax of #clock-time or #offset-time should be used throughout the subtitle document."
... I read the comment from the commenter on what this means to use the same syntax.
... I tried to be more specific here in this revision by stating that: "A Document Instance should use either the offset-time or offset-time syntax for all time expressions used therein."
... Oh I see there's a typo here. This should read: "A Document Instance should use either the clock-time or offset-time syntax for all time expressions used therein."
... This is just to make it easier for instance documents to read.

nigel: I'm not sure how this addresses Andreas's question.

pal: They would all be examples of clock time syntax so they would be the same syntax.
... The revised text tries to be more specific.

nigel: As written this doesn't require that all time expressions must use the same syntax throughout the document.
... How about:
... "All time expressions within a Document Instance should use the same syntax, either clock-time or offset-time."

pal: Looks fine to me.
... I'm making the change and will make it Pending after committing.
... LC-2974 #length-cell
... #length-cell was forbidden but that conflicted with ebutts:linePadding which uses length-cell. So I've updated the
... constraint to make an exception for ebutts:linePadding.
... Marking as pending.
... LC-2980 default Region
... I've now made this a defined term pointing back to TTML1's definition.
... Marking as pending.
... LC-2981 Reference fonts
... This is text we hadn't touched for a while so it prompted me to go into this deeper. The intent was really not so much
... to mandate a processor behaviour but to ensure that the document is authored using those reference fonts when applicable.
... Of course a presentation processor can choose whether or not to use those reference fonts.
... I've pulled out Reference Fonts from the HRM - the HRM is now only about document constraints.
... I've moved Reference Fonts to the Text Profile sections.
... This reads as a requirement on the authored document.

nigel: This doesn't actually require that text doesn't overflow the region even when laid out using the reference font metrics.

pal: That's an authorial decision - if the author wants to overflow that choice is available.
... The next step if we want to be stronger is to make it a normative requirement on presentation processors.
... That's the comment.

nigel: I'd made the same misreading.

pal: It's not entirely clear in CFF-TT. An implementor reading this and implementing a presentation processor would be
... ill-advised not to use the metrics of the reference fonts because otherwise the document will not look like the author
... intended.

nigel: The problem is that any test written against this spec text would have nothing to check.

pal: If there's any doubt here then I'd rather make it a presentation processor requirement to make it crystal clear and adopt
... what the commenter suggested, and be prescriptive on processors.

nigel: I think we need a good review period on this since it could be viewed as a substantive change by some readers.

pal: I'll make the edit as suggested by the commenter and set it to pending, and we can discuss it next week.
... For all the other ones that are marked pending can we ask the group to review and we can mark any with no comments
... as resolved.

nigel: I think so, yes.

Incoming Liaisons

nigel: We have an MPEG liaison and one from ITU-R Working Party 6B

http://www.itu.int/en/ITU-R/study-groups/rsg6/rwp6b/Pages/default.aspx

nigel: ITU has requested that IMSC be submitted before 10th November. Any objections to submitting it for information.

RESOLUTION: We will submit IMSC 1 to ITU-R Working Party 6B for information by 10th November.

<scribe> ACTION: nigel submit IMSC 1 to ITU-R Working Party 6B by 10th November [recorded in http://www.w3.org/2014/11/06-tt-minutes.html#action02]

<trackbot> Created ACTION-349 - Submit imsc 1 to itu-r working party 6b by 10th november [on Nigel Megitt - due 2014-11-13].

pal: I can assist if need be.

Action Items

action-341?

<trackbot> action-341 -- Nigel Megitt to Check with the mpeg folk if a wg note would be acceptable. -- due 2014-10-30 -- PENDINGREVIEW

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/actions/341

close action-341

<trackbot> Closed action-341.

action-342?

<trackbot> action-342 -- Nigel Megitt to Add change proposals to the tpac agenda for tuesday -- due 2014-10-30 -- PENDINGREVIEW

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/actions/342

close action-342

<trackbot> Closed action-342.

Issues raised

issue-351?

<trackbot> issue-351 -- Update IANA registration for TTML2 -- raised

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/issues/351

reopen issue-351

<trackbot> Re-opened issue-351.

issue-352?

<trackbot> issue-352 -- Add Media Registration Annex -- raised

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/issues/352

reopen issue-352

<trackbot> Re-opened issue-352.

issue-353?

<trackbot> issue-353 -- Normatively define short code processorProfiles parameter -- raised

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/issues/353

reopen issue-353

<trackbot> Re-opened issue-353.

Future meetings

nigel: I will make the November meetings all 1 hour unless anyone tells me they think we need longer (where 'anyone' includes me!)
... thank you everyone, let's meet in a week.
... [adjourns meeting]

Summary of Action Items

[NEW] ACTION: nigel submit IMSC 1 to ITU-R Working Party 6B by 10th November [recorded in http://www.w3.org/2014/11/06-tt-minutes.html#action02]
[NEW] ACTION: pal Thank DVB for input re LC-2983 [recorded in http://www.w3.org/2014/11/06-tt-minutes.html#action01]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.138 (CVS log)
$Date: 2014/11/06 16:13:01 $