W3C home > Mailing lists > Public > public-web-and-tv@w3.org > May 2014

[Minutes] Web & TV IG call - 28th May 2014

From: Daniel Davis <ddavis@w3.org>
Date: Wed, 28 May 2014 23:40:36 +0900
Message-ID: <5385F564.9000605@w3.org>
To: "public-web-and-tv@w3.org" <public-web-and-tv@w3.org>
CC: Mark Sadecki <mark@w3.org>
Hi all,

Here are the minutes from today's Web and TV Interest Group call:

Also pasted as text below.

With regards,


                               - DRAFT -

                           Web and TV IG call

28 May 2014



   See also: [3]IRC log

      [3] http://www.w3.org/2014/05/28-webtv-irc


          Bin, Alex, Kaz, Yosuke, Daniel, Paul, JC, Giri, Wu

          Giuseppe, Mark_Vickers




     * [4]Topics
         1. [5]Use-case gathering
         2. [6]Audio Fingerprinting (watermarking) - Daniel
         3. [7]Media playback adjustment - Daniel
         4. [8]Media stream synchronization
         5. [9]Triggered Interactive Overlay
         6. [10]Accessibility use case extensions
     * [11]Summary of Action Items

   <ddavis> scribenick: ddavis

Use-case gathering

   <yosuke> [12]http://www.w3.org/2011/webtv/wiki/New_Ideas

     [12] http://www.w3.org/2011/webtv/wiki/New_Ideas

   yosuke: So there are some use cases from IG members and several
   concerning accessibility.
   ... So going through the wiki page...
   ... First use-case is audio fingerprinting.

   <kaz> scribenick: kaz

Audio Fingerprinting (watermarking) - Daniel

   <inserted> [13]Audio Fingerprinting


   ddavis: ATSC asked about proposal on this
   ... Facebook has announced they're adding watermarking
   capability to their mobile apps

   <gmandyam> Just to clarify: The ATSC put out a
   call-for-proposals for watermarking video/audio with an
   embedded low rate data channel (around 1 kbps), where the user
   device could extract information in the broadcast feed to
   retrieve additional data related to the content being played.

   paul: uncertain about what to do...

   ddavis: would add clarification

   paul: great

   gmandyam: video watermark and audio watermark
   ... audio fingerprinting for location identification, etc.
   ... think video watermarking fits the description here

   paul: what do you mean by "video watermarking" and "audio
   ... microphone to content recognition

   ddavis: maybe there are two separate use cases here

   yosuke: agree

   ddavis: would split them then

   jcverdie: concerned about patents

   ddavis: thought about that...

   paul: this is more about APIs

   ddavis: need to investigate

   paul: EME also has APIs for content protection
   ... may be related to patents

   ddavis: EME itself is an exception
   ... would investigate that point

Media playback adjustment

   <inserted> [14]Media Playback Adjustment


   ddavis: adjust media playback speed, etc.

   <PaulHiggs> <video> can change speed

   <PaulHiggs> but audio is likely to be muted

   yosuke: thinks this feature is important for broadcasting
   ... any other questions?

   aldafu: playback rate for video
   ... possible to do it

   ddavis: but audio may be muted

   <ddavis> scribenick: ddavis

   kaz: When we investigate this, maybe we can consider extending
   it, e.g. for Japanese videos - if we try to playback at a
   higher speed, the player can play the video and sound faster
   and the speech sound is played back at the original frequency.
   ... So there's a special speech playback technology. The sound
   is not muted and the sound quality is not changed.

   yosuke: I think modern VCRs have similar functions.
   ... So when we playback with double speed the voice may be
   high-pitched, which we don't want.

   <PaulHiggs> how does the re-sync occur?

   <aldafu> ddavis: check

     [15] http://www.w3.org/2010/05/video/mediaevents.html

   yosuke: If the HTML5 media element doesn't have that function,
   it's worth considering.

   <aldafu> ddavis, you can actually control playback rate and
   audio goes along for me

   <aldafu> ddavis, not really, it becomes pretty warped

   <kaz> [some smart recorder has a capability of preserver the
   audio pitch]

Media stream synchronization

   ddavis: Next use case is about media sychronisation. For
   example, two people watching the same content on two devices
   (e.g. next to each other on the train) would like to have the
   videos synchronised so they can laugh at the same jokes at the
   same time.

   <PaulHiggs> there are several industry initiatives on screen

   <inserted> [16]Media Stream Synchronization


   yosuke: Using watermarking can take time so it may not work.

   PaulHiggs: There are a few other industry organisations looking
   at synchronising streams - AKA screen synchronisation or
   companion device synchronisation.

   yosuke: This use case has some new perspective.
   ... The broadcasting industry is syncing a broadcast stream and
   video, whereas this use case is about syncing two videos.

   PaulHiggs: I read a couple of things in this - there's
   multi-camera angle synchronisation and multi-video
   ... It could be 2.5 use cases. The multi-angle one could be on
   the same device or separate devices.

   <scribe> ACTION: ddavis to clarify use case 1 description
   [recorded in

   <trackbot> Created ACTION-194 - Clarify use case 1 description
   [on Daniel Davis - due 2014-06-04].

   <scribe> ACTION: ddavis to check current ability to change
   playback rate for HTML media element. [recorded in

   <trackbot> Created ACTION-195 - Check current ability to change
   playback rate for html media element. [on Daniel Davis - due

   <scribe> ACTION: ddavis to split synchronisation use case (#4)
   [recorded in

   <trackbot> Created ACTION-196 - Split synchronisation use case
   (#4) [on Daniel Davis - due 2014-06-04].

Triggered Interactive Overlay

   <inserted> [20]Triggered Interactive Overlay UC


   yosuke: Bin, could you explain use case 4 please?

   Bin_Hu: When you watch TV, based on the service provider you'll
   be provided with an overlay showing the promotion of another
   channel, for example.
   ... The content is triggered by content in the main stream.
   ... There are a few requirements that can be extracted from
   ... So the trigger could be within the stream, or it could be
   within the platform and associated with a channel.
   ... Also, you may be viewing a baseball game, for example, and
   there could be a trigger that shows an overlay.
   ... This overlay could show a different team.
   ... These overlays must be valid during that 15 or 30 seconds.
   ... Or it could be triggered by a particular actor in the

   <PaulHiggs> can you clarify why this only applies to a "Hybrid

   Bin_Hu: This trigger must be technology-agnostic.

   <PaulHiggs> HbbTV and MPEG DASH already provide "standardized"

   Bin_Hu: HTML5 may be enhanced by added event types to support
   these different triggers.
   ... On the other hand, in HTML5 there may be some other ways,
   specifically in specs for the TV industry.

   yosuke: These trigger event types may be related to the Media
   In-Band Resource Community Group.
   ... They're looking at MPEG events and tracks

   Bin_Hu: Right, it may be part of that or it could be something
   for another group to look into.

   PaulHiggs: In the use case you say this applies to a hybrid
   STB. Any reason why it has to be a hybrid STB?

   Bin_Hu: That's just an example. Maybe IP STB or other is also

   jcverdie: I'm confused about the difference about your use case
   and a basic web app to get web notifications from the server.

   Bin_Hu: For example with sports, you may have a trigger engine
   within the STB, e.g. at 7:00pm, so the user may be offered
   other content.
   ... But how to enable those events on the web platform is the
   ... How to create the same experience on top of playing content
   (e.g. Channel 5)?
   ... The native platform is able to use such trigger engines.

   yosuke: Any other comments?
   ... Bin, I think this is interesting so I'd like to do some gap
   analysis of this use case and the Media In-Band Resource CG.
   ... I'll do that quickly and put the results on the mailing

   Bin_Hu: Great, thank you.

   kaz: I'd also like to join the gap analysis work because
   there's a slight possibility of MMI work being related.

   <PaulHiggs> what is MMI?

Accessibility use case extensions

   yosuke: Do we have any of the accessibility people here today?

   kaz: No but we can invite them for the next time.

   <PaulHiggs> is this "accessability" for those with disabled?

   ddavis: These are additional parts to the use cases from the
   first round.

   kaz: These are not necessarily brand new use cases. We can
   start by briefly checking the use cases additions that have
   been submitted.

   yosuke: So you suggest we walk through them quickly?

   kaz: Yes, some of them.
   ... Maybe we can just pick up the accessibility portion from
   some of them.

   yosuke: Do you have any suggestions about which to look at?

   kaz: Mark Sadecki (W3C) and John Foliot (member) created these.
   John considered media accessibility requirements so let's look
   at UC 8.

   <kaz> [21]Accessibility Extension Use Case 8


   yosuke: Use case is Download and Go.
   ... The original use case is when the user downloads content on
   to a tablet to watch later.
   ... Accessibility requirement is that whenever a media file is
   downloaded, all related resources (captions, etc.) must also be
   available for download.
   ... The download of the supplemental resources should maintain
   the hierarchy.
   ... The file directory structure is preserved in the download.
   ... For example, putting all files in a zip or cab file.

   kaz: So this means we have to think about how to pack all
   related content.
   ... This is similar to ePub format for digital books.

   yosuke: The reality of the industry is that the recorder splits
   the information from the original stream.
   ... I think we can suggest something to the TV industry.

   kaz: We could also think about media synchronisation for this
   purpose as well.
   ... Not only in-band information but also out-band information
   as well.

   yosuke: I think there are two ways to deal with these
   ... One way is to improve these use cases.
   ... But to me, these sound like more fundamental requirements.
   ... It sounds like something similar to GGIE that is now
   talking about nodes between original creators and consumers and
   that content should not change between them.
   ... We could create a note or guideline about essential
   requirements for web and TV.
   ... Which way do you think fits into W3C's way of working best?

   kaz: Maybe we could start with a note or guidelines and if
   needed pick up additional requirements or use cases later.

   ddavis: So we can get general, more common accessibility issues
   as a note or set of guidelines, and then particular ones may be
   better as a new use case.

   yosuke: People form the industry can reference what kind of
   requirements they need to satisfy by looking at our guidelines.
   For something more specific we can have a separate use case.
   ... So Kaz or Daniel, could you feed this back to the
   accessibility group?

   ddavis: Yes

   yosuke: Should we invite them to this call?

   kaz: Yes, we can do that.

   yosuke: Should the IG do something before the next call?
   ... I can classify their requirements into general ones and
   specific ones.

   ddavis: That sounds very useful to me.

   kaz: we can also send a call for participation on this topic.

   <gmandyam> Giri signing off - thanks.

   ddavis: The deadline for use case submissions is the end of
   this week, right?

   kaz: Yes, but we can polish use cases after that.

   yosuke: So we've gone through all the use cases.

   <scribe> ACTION: Yosuke to classify the accessibility
   requirements into general ones and specific ones. [recorded in

   <trackbot> Created ACTION-197 - Classify the accessibility
   requirements into general ones and specific ones. [on Yosuke
   Funahashi - due 2014-06-04].

   ddavis: With 4k content starting to increase (e.g. Netflix) is
   this something that can affect our use cases? Do we need new

   yosuke: In Japan broadcasters have started looking at caption
   for 4k but it's still very early stage.
   ... I'm not sure the caption people have considered 4k. It may
   be better to ask them if there are gaps or problems.

   ddavis: OK, I'll ask them.

   yosuke: During the last TPAC the work of the Community Group
   was migrated into the Working Group.

   ddavis: I can speak to the Timed Text Working Group

   <scribe> ACTION: ddavis to ask Timed Text WG about 4k affecting
   captioning. [recorded in

   <trackbot> Created ACTION-198 - Ask timed text wg about 4k
   affecting captioning. [on Daniel Davis - due 2014-06-04].

   yosuke: There may be other non-captioning issues with 4K.
   ... We can take a step-by-step approach, looking at captioning
   ... Another alternative is to also discuss other issues within
   the IG.
   ... What do you think?

   kaz: I think step-by-step approach is better.

   yosuke: Let's go with a step-by-step approach.

   PaulHiggs: If we're going to start looking at future
   technologies, there's all kind of things about gesture and
   speech-based input/control.
   ... Are you expecting use cases on those?

   PaulHiggs: So we could focus on 4K but then we don't look at
   something else.

   yosuke: Any topic that IG members or the industry is interested
   in can be the topic for these use cases.
   ... For example, Netflix is bringing 4K content so it's worth
   looking at.
   ... Japanese industry will also start broadcasting 4k trials
   during this year's world cup. So it's not a far-future
   ... If there's something that can affect web users, it could be
   a relevant topic.

   ddavis: There's also a requirement that any use case is fine as
   long as somebody has ownership of it.

   PaulHiggs: Should Netflix be bringing this to the discussion?

   yosuke: What about asking the stakeholders to submit issues or
   ... There are some research facilities dealing with 4k - we can
   ask them about issues and web standards.

   kaz: You can also bring your ideas to the wiki.

   PaulHiggs: What I'm trying to say is let's solve a problem
   where there is one.
   ... I don't think we can answer what they need - only they can
   answer that.

   yosuke: Paul's point is good, so how about creating an
   invitation or questionnaire for 4k stakeholders?
   ... We can create the text and ask stakeholders, sharing the
   result with the IG.
   ... Based on that we can decide what to do next.

   <scribe> ACTION: ddavis and yosuke to create questionnaire for
   4k stakeholders about web standards issues. [recorded in

   <trackbot> Created ACTION-199 - And yosuke to create
   questionnaire for 4k stakeholders about web standards issues.
   [on Daniel Davis - due 2014-06-04].

   yosuke: Thank you all. Let's do follow-up work on the mailing

   <kaz> [ adjourned ]

Summary of Action Items

   [NEW] ACTION: ddavis and yosuke to create questionnaire for 4k
   stakeholders about web standards issues. [recorded in
   [NEW] ACTION: ddavis to ask Timed Text WG about 4k affecting
   captioning. [recorded in
   [NEW] ACTION: ddavis to check current ability to change
   playback rate for HTML media element. [recorded in
   [NEW] ACTION: ddavis to clarify use case 1 description
   [recorded in
   [NEW] ACTION: ddavis to split synchronisation use case (#4)
   [recorded in
   [NEW] ACTION: Yosuke to classify the accessibility requirements
   into general ones and specific ones. [recorded in

   [End of minutes]
Received on Wednesday, 28 May 2014 14:41:16 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:57:21 UTC