See also: IRC log
yosuke: approve his participation in this call
tta: TTA is a telecom tech association in Korea
kaz: we can invite him to this specific call based on Yosuke's approval
yosuke: also simple
liaison with them would be good
... so that we can invite them to the IG meeting at TPAC 2014
<yosuke> https://www.w3.org/2011/webtv/wiki/New_Ideas#Use_Cases
yosuke: walked through the new use case descriptions
<kaz> UC 2-1
<kaz> UC 2-2
yosuke: let's begin with uc2-1
... added "Gaps" section
... three gaps
... no standard for syntax for Web APIs yet
... next feature for accessing hash generators
... hash modules for client devices
... may be proprietary algorithm
... we don't have I/F for that
... the third point is in-browser capability to generate hashes
... those are three potential gaps
... Daniel is the submitter of this UC
... so would like Daniel's opinion
ddavis: possibly the
last one might not be implemented in browsers
... the first one could be important
... the second one could be related to EME
... could be used for not only fingerprinting but also other services
yosuke: so you mean we should concentrate on the second point?
ddavis: I think they all have potential but maybe "Feature to access proprietary hash generators." is the one that has most potential
yosuke: agree
ddavis: It could have more impact as it looks like it could be used for other algorithms as well, not just audio fingerprinting.
yosuke: if we carefully
design external module for EME, we can reuse it for audio
fingerprinting as well
... a generic API would be useful for both the use cases
... helpful for Web developers
paul: But I think watermarking is used when listening to content in a separate device.
yosuke: you can use
that module for multiple devices
... not depending on concrete devices
paul: watermarking depends on device architecture, doesn't it?
yosuke: could be implemented independent from devices
paul: my understanding
was watermarking was high frequency codes
... decoded by client devices
CyrilRa: You could also use a client-server structure for watermarking
paul: evidence-based and information-based
yosuke: we can clarify
the models of watermarking
... might need help of proprietary modules
... if we create some new I/F, we should not define different I/F for fingerprinting and watermarking
... we should try to have generic architecture for both purposes
... that's my opinion
... make sense?
paul: could do that
ddavis: good to hear broader range of opinions
yosuke: this kind of modular model could be problematic
... users might be able to access the core module, e.g., EME
... but this time, even if we have proprietary modules, it's just generating hash data
... not directly related to the content
... so this modular system could be safer
paul: client-server mechanism?
ddavis: mainly thinking of client-side browser
paul: we should recommend using server-side approach
yosuke: basically this modular-based approach is safer
... but if there is anybody who cares, should we ask TAG or HTML-WG?
ddavis: the media tf of the HTML-WG
paul: thinking about something like XML HTTP Request
... local microphone captures sounds for apps
ddavis: APIs expose modules
... for external servers
paul: implementation could use XYZ server
... presentation generated by themselves (local-side)
ddavis: one more HTTP Request
yosuke: that sounds like the first gap
<yosuke> Client-server web interface (e.g. RESTful API) for fingerprinting services, both queries and returned results.
ddavis: yes
yosuke: interesting
... think we should ask actual audio fingerprinting providers for opinions
... good I/F between existing fingerprinting systems and Web services
ddavis: good idea
yosuke: after this
call, the co-Chars and the Team Contacts should contact actual
fingerprinting providers
... also should talk with related W3C groups, e.g., HTML WG Media TF
... what do you think?
(no objections)
<kaz> ACTION: yosuke to work with Kaz/Daniel and contact wartermarking/fingerprinting providers [recorded in http://www.w3.org/2014/09/03-webtv-minutes.html#action01]
<trackbot> Created ACTION-206 - Work with kaz/daniel and contact wartermarking/fingerprinting providers [on Yosuke Funahashi - due 2014-09-10].
<kaz> ACTION: yosuke to talk with related W3C groups about possible architecture for watermarking/fingerprinting [recorded in http://www.w3.org/2014/09/03-webtv-minutes.html#action02]
<trackbot> Created ACTION-207 - Talk with related w3c groups about possible architecture for watermarking/fingerprinting [on Yosuke Funahashi - due 2014-09-10].
yosuke: maybe not only the HTMLWG but also SysApps WG
ddavis: maybe WebApps and Audio as well
kaz: maybe VBWG as well
yosuke: yes
<kaz> UC 2-3
<ddavis> scribenick: ddavis
kaz: Identical stream
synchronization gaps have been identified as:
... "No ability to use existing synchronization data within a
stream. "
... "No way to accurately synchronize the presentation of media
elements on multiple devices (with the same media
source)."
... We thought this use case is for related multiple streams as
well (the next use case).
... There is a lack of synchronization mechanism for multiple
streams (identical or not)
<kaz> scribenick: kaz
yosuke: media capability of HTML5
doesn't have accurate synchronization mechanism
... we should dig into that issue
... from a fundamental viewpoint
... not only for TV industry but also various Web apps
... during the upcoming IG f2f meeting at TPAC, we'll talk
about a possible CG for media stream synchronization
... already have second screen CG, but no group for media
synchronization in general
... at the same time, need to know synchronization mechanism
outside W3C
... e.g., HbbTV, Hybridcast (IPTVF-Japan), etc., have their own
approach
... so would suggest we talk about generating a possible CG or
a TF within this IG for that topic
... what do you think?
kaz: myself like creating a TF
within this IG
... but would ask all for opinions
yosuke: several possible
solutions
... a TF within this IG
... a new CG
... or a TF within the HTML-WG
... which would be the best for us?
<bill> I agree - leveraging and aligning other standards is especially relevant. ALso note AllPlay.
yosuke: would talk about that
during TPAC
... CG has a strong point and a weak point
... we should talk about those points carefully
kaz: so you want to add this topic to the TPAC f2f agenda?
<Paul_Higgs> sorry, I have to leave now
yosuke: we can even ask related
groups for opinions before TPAC
... might be going to hold a joint session with other groups,
e.g., the HTML-WG
... because this is related to media element
... any objections?
(no objections)
<Geun_Hyung_Kim> I
-> https://www.w3.org/2011/webtv/wiki/New_Ideas#UC2-4_Related_Media_Stream_Synchronization UC 2-4
yosuke: this has some unique gaps from the previous one
<ddavis> scribenick: ddavis
kaz: This use case is related to
the previous one.
... Gaps identified so far:
... "No ability to use existing synchronization data within a
stream."
... "No way to accurately synchronize multiple media elements
in a single document (with different but related media
sources)."
... "No way to deal with timestamp differences in related media
streams. Note that the timestamp formats may differ in the
streams, e.g. if one is 24 fps and one is 60 fps."
... This third gap is quite unique and different to the
previous use case.
<kaz> scribenick: kaz
yosuke: yes, the third one is
most important
... how to deal with synchronization?
... would note that and move on
ddavis: time stamp format question
yosuke: current work doesn't
include this kind of capability
... we can ask their interest
kaz: who do you mean by they?
yosuke: media in-band CG and TTML
WG
... synchronization between text track and media track
yosuke: any other questions?
(nothing)
yosuke: let's move forward
-> https://www.w3.org/2011/webtv/wiki/New_Ideas#UC2-5_Triggered_Interactive_Overlay UC 2-5
ddavis: overlay on the top of TV
and overlay itself
... lot of things done
... HTML dialog could be used
... and this use case should focus on TV use cases
... when something changes, e.g., commercial or channels
... the UC talks about binding and content
... no way for binding outside TV area
... label element
... no binding between elements
[[
Extended media events to notify of things like section and scene changes (e.g. commercial break) and change of media source (e.g. change of channel).
No standard to bind video content with in-browser objects.
]]
bin: the gap is pretty much
inline with the requirement of the TV Control CG
... fine-tune events
... for interactive overlay
... based on pre-defined schedule
bin: conditions of triggering
events
... dispatched based on different conditions
... the gap analysis is inline with what we have been
discussing
... in-band method and out-band method
yosuke: sounds like it's relevant
to bring this UC to the TV Control API CG
... any other comments?
(nothing)
-> https://www.w3.org/2011/webtv/wiki/New_Ideas#UC2-6_Clean_Audio UC 2-6
yosuke: clarified two gaps
[[
No way to accurately synchronize multiple media elements in a single document (with different but related media sources).
No way to equalize content-protected (e.g. with EME) audio tracks.
]]
yosuke: even service providers,
e.g., Netflix and YouTube, provide clean audio in addition to
ordinary track
... synchronization mechanism is needed
... we should work on that
... we can deal with this kind of important feature
(synchronization) within one new group/tf
... the second one
... using Web APIs, we can equalize ordinary audio track
... using Web Audio APIs
... but impossible for users to change the tone
... we should ask the Accessibility TF guys to deal with
this
... also with the HTML Media TF
... ask them about their opinions on the gaps
... what do you think?
ddavis: sounds like good
idea
... at least good starting point
... asking the a11y tf and the media tf
yosuke: we can talk about this
topic and other accessibility issues during the f2f meeting at
TPAC
... maybe we can have a brief talk during our IG meeting with
them (a11y tf)
... also talk about how to deal with this use cases
kaz: during the a11y tf's meeting?
yosuke: accessibility tf should
be going to hold their meeting at TPAC
... and so I think we should invite them to our IG meeting
kaz: so we'll invite the A11Y TF
guys to the Web&TV IG meeting on Monday
... and some of the Web&TV IG participants will visit the
A11Y TF on Thu/Fri
... ok
... I'll talk with Janina about the possible slots
yosuke: tx
... on Monday we'll have a joint session with the A11Y guys
during the IG meeting
... the IG meeting will be held on Monday
... their meeting will be held on Thu/Fri
... we should announce the agenda to the IG meeting
participants
... the sooner, the better
kaz: +1
<scribe> ACTION: kaz to talk with Janina about the joint session slot during TPAC 2014 [recorded in http://www.w3.org/2014/09/03-webtv-minutes.html#action03]
<trackbot> Created ACTION-208 - Talk with janina about the joint session slot during tpac 2014 [on Kazuyuki Ashimura - due 2014-09-10].
yosuke: any other questions?
(nothing)
yosuke: will summarize today's
discussion and update the wiki based on the discussion
... also will draft agenda for TPAC f2f
... will put it on the wiki
... and share the information with the co-Chairs and the IG
participants
... we've started to think about the agenda, so need to
finalize the possible joint meeting with the A11Y TF
yosuke: in 2 weeks, i.e., Sep.
17th
... will talk about TPAC agenda during the call
... might be going to have an unconference style
kaz: the TV Control CG will use the morning sessions
bin: will finalize it soon
<Zakim> ddavis, you wanted to remind about second screen questionnaire
<ddavis> https://www.w3.org/2002/09/wbs/33280/secondscreen/ (Member-only)
ddavis: please ask your AC Rep. to respond the questionnaire by Sep. 22
yosuke: any other business?
skim13: given we'll talk about the TPAC agenda, can we invite TTA guys to the next call on 17th as well?
yosuke: would like to talk with
the other co-Chairs as well first
... would some more information from you and TTA
... we should handle this carefully
[ adrjouned ]