09-mediafrag-minutes.html 45 KB

Raw Blame History Permalink

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
    "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">

<html lang='en' xmlns="http://www.w3.org/1999/xhtml" xml:lang="en">
<head>
  <meta name="generator" content=
  "HTML Tidy for Linux/x86 (vers 12 April 2005), see www.w3.org" />

  <title>Media Fragments Working Group Teleconference -- 09 Dec
  2008</title>
  <link type="text/css" rel="STYLESHEET" href=
  "http://www.w3.org/StyleSheets/base.css" />
  <link type="text/css" rel="STYLESHEET" href=
  "http://www.w3.org/StyleSheets/public.css" />
  <link type="text/css" rel="STYLESHEET" href=
  "http://www.w3.org/2004/02/minutes-style.css" />
  <meta content="Media Fragments Working Group Teleconference"
  name="Title" />
  <meta content="text/html; charset=utf-8" http-equiv=
  "Content-Type" />
</head>

<body>
  <p><a href="http://www.w3.org/"><img src=
  "http://www.w3.org/Icons/w3c_home" alt="W3C" border="0" height=
  "48" width="72" /></a></p>

  <h1>- DRAFT -</h1>

  <h1>Media Fragments Working Group Teleconference</h1>

  <h2>09 Dec 2008</h2>

  <p><a href=
  'http://www.w3.org/2008/WebVideo/Fragments/wiki/SecondF2FAgenda'>Agenda</a></p>

  <p>See also: <a href=
  "http://www.w3.org/2008/12/09-mediafrag-irc">IRC log</a></p>

  <h2><a name="attendees" id="attendees">Attendees</a></h2>

  <div class="intro">
    <dl>
      <dt>Present</dt>

      <dd>Yves, Frank, Davy, Erik, Raphael, Tom,
      Silvia_(remote)</dd>

      <dt>Regrets</dt>

      <dt>Chair</dt>

      <dd>Erik, Raphael</dd>

      <dt>Scribe</dt>

      <dd>raphael, erik</dd>
    </dl>
  </div>

  <h2>Contents</h2>

  <ul>
    <li>
      <a href="#agenda">Topics</a>

      <ol>
        <li><a href="#item01">1. Admin</a></li>

        <li><a href="#item02">2. Discussion Existing
        Technologies</a></li>

        <li><a href="#item03">3. Define Types of
        Addressing</a></li>

        <li><a href="#item04">4. Implementation Issues (protocol
        &amp; caching)</a></li>

        <li><a href="#item05">5. Joint Session with Media
        Annotations WG</a></li>

        <li><a href="#item06">joint meeting MF &amp; MA</a></li>
      </ol>
    </li>

    <li><a href="#ActionSummary">Summary of Action Items</a></li>
  </ul>
  <hr />

  <div class="meeting">
    <p class='phone'>&nbsp;</p>

    <p class='phone'>&nbsp;</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Date: 09 December
    2008</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; scribenick:
    raphael</p>

    <h3 id="item01">1. Admin</h3>

    <p class='phone'>I would like we talk about the composition of
    the group in order to know if more people/companies are about
    to join</p>

    <p class='phone'>YouTube/Google Video: Ken Harrenstien is more
    interested in Media Annotations</p>

    <p class='phone'><cite>scribe:</cite> it would be interested to
    have someone that has implemented fragments access</p><a name=
    "action01" id="action01"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> Yves to find out with Philippe who
    from Google would be interested to join [recorded in <a href=
    "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action01">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action01</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Created ACTION-17
    - Find out with Philippe who from Google would be interested to
    join [on Yves Lafon - due 2008-12-16].</p><a name="action02"
    id="action02"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> Raphael to see with Marie Claire who
    from Daily Motion can join [recorded in <a href=
    "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action02">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action02</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Sorry, couldn't
    find user - Raphael</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; ACTION-2 on Troncy</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; ACTION-2 due December
    16 2008</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; ACTION-2 Set up a
    questionary for seond MediaFrag F2F in Gent (8. and 9. Dec) due
    date now December 16 2008</p>

    <p class='phone'>Frank (Canon): it would be difficult for Canon
    to join in 2009</p>

    <p class='phone'><cite>Adobe:</cite> Larry Masinter answered,
    he has not yet someone to nominate in the group, but Adobe
    supports strongly this group</p><a name="action03" id=
    "action03"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> Troncy to check with Karen about Blinx
    joining or not W3C and Colm the WG [recorded in <a href=
    "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action03">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action03</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Created ACTION-19
    - Check with Karen about Blinx joining or not W3C and Colm the
    WG [on Raphaël Troncy - due 2008-12-16].</p><a name="action04"
    id="action04"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> Michael to check with Wolfgang whether
    he is still interested in this WG [recorded in <a href=
    "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action04">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action04</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Created ACTION-20
    - Check with Wolfgang whether he is still interested in this WG
    [on Michael Hausenblas - due 2008-12-16].</p>

    <p class='irc'>&lt;<cite>nessy</cite>&gt; I'm idling</p>

    <p class='irc'>&lt;<cite>nessy</cite>&gt; will have go for 2
    hours, but back then</p><a name="action05" id="action05"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> Erik to check with Philippe the status
    of Cisco (Paul Bosso), Apple (Dave Singer or Eric Carlson)
    [recorded in <a href=
    "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action05">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action05</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Created ACTION-21
    - Check with Philippe the status of Cisco (Paul Bosso), Apple
    (Dave Singer or Eric Carlson) [on Erik Mannens - due
    2008-12-16].</p><a name="action06" id="action06"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> Raphael to check with Karen the status
    of Fox Interactive, if they could have an interest in the group
    [recorded in <a href=
    "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action06">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action06</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Sorry, couldn't
    find user - Raphael</p><a name="action07" id="action07"></a>

    <p class='irc'>&lt;<cite>Yves</cite>&gt;
    <strong>ACTION:</strong> Rapha�l to check with Karen the status
    of Fox Interactive, if they could have an interest in the group
    [recorded in <a href=
    "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action07">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action07</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Sorry, couldn't
    find user - Rapha�l</p>

    <p class='phone'><cite>Erik:</cite> should we have a stronger
    liaison with HTML5?</p>

    <p class='phone'><cite>Yves:</cite> we have work to do, it is
    good to keep contact, but we could ask more feedback when we
    have better documents<br />
    ... same for browser vendors</p>

    <h3 id="item02">2. Discussion Existing Technologies</h3>

    <p class='phone'>On the wiki: <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/wiki/Existing_Technologies_Survey">
    http://www.w3.org/2008/WebVideo/Fragments/wiki/Existing_Technologies_Survey</a></p>

    <p class='phone'>Tom (IBBT) going through the wiki</p>

    <p class='phone'>Presentation also available at: <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/meetings/2008-12-09-f2f_ghent/IBBT-State_of_the_art.pptx">
    http://www.w3.org/2008/WebVideo/Fragments/meetings/2008-12-09-f2f_ghent/IBBT-State_of_the_art.pptx</a></p>

    <p class='phone'><cite>Tom:</cite> first explain what SMIL can
    do (Jack will be here later today and tomorrow)<br />
    ... MPEG-7 (see slide 3)</p>

    <p class='phone'><cite>Raphael:</cite> should we discuss the
    format for representing the time point?</p>

    <p class='phone'><cite>Yves:</cite> you can adopt the ISO Dates
    one, the XML Schema one<br />
    ... the MPEG-7 one is based on XML Schema, minus the Time Zone,
    but adding the frame number</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt;
    ttp://www.w3.org/2002/ws/databinding/edcopy/report/all.html</p>

    <p class='phone'><cite>Better:</cite> <a href=
    "http://www.w3.org/2002/ws/databinding/edcopy/report/all.html">http://www.w3.org/2002/ws/databinding/edcopy/report/all.html</a></p>

    <p class='phone'><cite>Yves:</cite> we should say we consider
    only time that is local to the media<br />
    ... so de don't care about time zones for example</p>

    <p class='phone'><cite>Tom:</cite> SVG has no temporal
    fragment<br />
    ... TimedText: it shows text at a given time<br />
    ... seems to have another format for representing time
    point<br />
    ... CMML derives from Annodex, it requires an off file that has
    been annotated</p>

    <p class='phone'><cite>Yves:</cite> Silvia points the problem
    of accessing a given frame, if it is not an I-Fram</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; (when referencing only
    time)</p>

    <p class='phone'><cite>Raphael:</cite> we can decide to always
    go to the previous I-Frame, that precedes a time point</p>

    <p class='phone'><cite>Tom:</cite> CMML specifies time with
    npt, smpte and clock</p>

    <p class='phone'><cite>Raphael:</cite> should we do the
    same?</p>

    <p class='phone'><cite>Tom:</cite> default seems to be npt</p>

    <p class='phone'><cite>Raphael:</cite> we come back to these
    questions when Sivlia is on the phone</p>

    <p class='phone'><cite>Tom:</cite> CMML/Annodex/TemporalURI has
    no spatial Fragment<br />
    ... can select Tracks (such as in a CD)<br />
    ... has the notion of naming a fragment and refer to this
    name</p>

    <p class='phone'><cite>Yves:</cite> is there an error in the
    named fragment example? Should the '/' be escaped for selecting
    the tracks 'a' and 'b' ?</p>

    <p class='phone'><cite>Tom:</cite> MPEG-21 has 4 different
    schemes (ffp, offset, mp, mask)<br />
    ... offset works in bytes range<br />
    ... mp scheme has the time dimension (npt, smpte, utc, mpeg-7)
    and the spatial dimension (polygon, rectangle, elipse)<br />
    ... mask is similar to kind of naming a fragment<br />
    ... HTML5 (see slide 5)<br />
    ... no support for fragmentation or time reference (like in
    SVG)<br />
    ... has Time Daatatypes: Date, Time, Date and Time, Time Zones
    (UTC: add a Z at the end; others: add time difference to UTC
    with + or -)<br />
    ... values come from XML Schema (perhaps with one small
    difference, since the seconds can be omitted)</p>

    <p class='phone'><cite>Raphael:</cite> go through the spatial
    fragments specifications (image maps, MPEG-7, SVG)</p>

    <p class='phone'><cite>Erik:</cite> how this technological
    survey be used ?</p>

    <p class='phone'><cite>Raphael:</cite> we will provide either
    informally or in the spec a mapping between the URI schem and
    these various XML syntaxes</p>

    <p class='phone'><cite>Yves:</cite> Since we want a URI scheme,
    we will not support everything we have seen, but the maximal
    possible subset</p>

    <p class='irc'>&lt;<cite>nessy</cite>&gt; back now</p>

    <p class='irc'>&lt;<cite>rtroncy</cite>&gt; Silvia, we have a
    number of questions for you :-)</p>

    <p class='phone'>Question 1: We reviewed <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/wiki/Existing_Technologies_Survey#CMML">
    http://www.w3.org/2008/WebVideo/Fragments/wiki/Existing_Technologies_Survey#CMML</a></p>

    <p class='phone'><cite>scribe:</cite> we wonder if there is not
    a mistake in the URI example, and if the '/' should not be
    escaped</p>

    <p class='phone'><cite>Silvia:</cite> first we use a '-' and
    then move to '/'</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; in the examples on
    CMML, the / after the ? should be escaped</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; =&gt; %2f</p>

    <p class='phone'><cite>Silvia:</cite> I think it is ok to have
    the '/' in the fragment ('#') but not for the query ('?')</p>

    <p class='phone'>Silvia will check whether there is a syntax
    error or not</p>

    <p class='phone'>Question 2: CMML covered 3 schemes for
    representing time point: npt, smpte and clock</p>

    <p class='phone'><cite>Raphael:</cite> should we do the
    same?</p>

    <p class='phone'><cite>Silvia:</cite> we wanted to be
    interoperable with all formats<br />
    ... in practice, people tend to use the 'npt' scheme<br />
    ... maybe it is better to talk with video professionals<br />
    ... they need to access the frame level<br />
    ... I think that for most use cases, the npt scheme is accurate
    enough<br />
    ... npt is the default scheme in CMML<br />
    ... don't confuse: ntp, the network time protocol (unix) and
    npt, what we are discussing<br />
    ... npt = normal playback time</p>

    <p class='phone'><cite>Raphael:</cite> Question 3: Frame
    access, should we always go to the last I-Frame that precedes
    the time point we want to access</p>

    <p class='phone'><cite>Silvia:</cite> depends on the what the
    codecs allows<br />
    ... with Theora, we jump to always to the previous
    I-Frame<br />
    ... we need to be accurate when we store the fragment (cache),
    it seems less important on the client side</p>

    <p class='phone'><cite>Davy:</cite> I agree with Silvia, we
    might want to provide some guidelines for some specific
    formats<br />
    ... we cannot define an algorithm that says that a time point
    corresponds to a particual frame for all encoding cases, it's
    not possible</p>

    <p class='phone'><cite>Silvia:</cite> we can say that previous
    I-Frame is accurate enough</p>

    <p class='phone'><cite>Raphael:</cite> what the cache will
    finally store?</p>

    <p class='phone'><cite>Silvia:</cite> Cache will store what the
    servers is serving, and recompose fragments based on bytes, not
    using the URI requested by the UA</p>

    <p class='irc'>&lt;<cite>davy</cite>&gt; <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/wiki/HTTP_implementation">
    http://www.w3.org/2008/WebVideo/Fragments/wiki/HTTP_implementation</a></p>

    <h3 id="item03">3. Define Types of Addressing</h3>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; zakim: mute me</p>

    <p class='phone'><a href=
    "http://www.w3.org/2008/WebVideo/Fragments/wiki/Types_of_Fragment_Addressing">
    http://www.w3.org/2008/WebVideo/Fragments/wiki/Types_of_Fragment_Addressing</a></p>

    <p class='phone'><cite>Erik:</cite> page prepared by Davy (with
    Guillaume input?)</p>

    <p class='phone'><cite>Davy:</cite> this page has origin from
    the list of issues we have discussed during our 1st face to
    face meeting<br />
    ... Track: whether a media format supports tracks or not
    depends on the Container format, but not the Coding format</p>

    <p class='phone'><cite>Frank:</cite> do you consider all the
    video quality level in one track? for example in a media
    adaptation use case</p>

    <p class='phone'><cite>Davy:</cite> I do not think that a
    different quality of the video is a fragment</p>

    <p class='phone'><cite>Raphael:</cite> discussion about what is
    the boudaries of the track definition</p>

    <p class='phone'><cite>Yves:</cite> examples such as multiple
    camera angles, multiple resolution of the same video in the
    same stream, audio languages, subtitles: are all of these
    tracks ?</p>

    <p class='phone'><cite>Silvia:</cite> the boundary should be
    what the encapsulation format exposes<br />
    ... or rather the container format</p>

    <p class='phone'><cite>Raphael:</cite> if the container format
    exposes the notion of tracks, we could address them, otherwise,
    we should NOT invent them<br />
    ... we look at the table</p>

    <p class='phone'><cite>Silvia:</cite> different camera angles
    can be seen as multiple video tracks<br />
    ... different resolution: encoding format does not work that
    way, they tend to provide different files<br />
    ... we should not worry about that now, can be dealt with
    later</p>

    <p class='phone'><cite>Davy:</cite> Temporal dimension: need to
    take into account the precision we can get in the time
    point<br />
    ... Spatial dimension: we cannot generally extract a region,
    not make yet a decision if we consider only rectangle regions
    or arbitrary shapes<br />
    ... Name dimension: again depends on the container format! For
    example, one can include a CMML or TimedText description in a
    MP4 or Ogg container</p>

    <p class='phone'><cite>Silvia:</cite> QuickTime has
    'QuickTimeText' that can be used to jump to a dvd chapter</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; cueranges</p>

    <p class='phone'><cite>Silvia:</cite> Flash has cueranges</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; <a href=
    "http://www.apple.com/quicktime/tutorials/texttracks.html">http://www.apple.com/quicktime/tutorials/texttracks.html</a></p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; ups,
    s/cueranges/cuepoints/</p>

    <p class='phone'><cite>Yves:</cite> <a href=
    "http://help.adobe.com/en_US/Soundbooth/2.0/WSA5A1DDFB-6BE2-4486-BE0C-A10CEEF119ADa.html">
    http://help.adobe.com/en_US/Soundbooth/2.0/WSA5A1DDFB-6BE2-4486-BE0C-A10CEEF119ADa.html</a>
    ?</p>

    <p class='phone'><cite>Davy:</cite> the table is not complete
    yet, for some format, I couldn't figure out what is possible or
    not<br />
    ... summary is that generally, the temporal dimension is not a
    problem<br />
    ... for the spatial dimension, this is more problematic!<br />
    ... a ROI can be extracted with H264, but this is not a crop,
    rather a decrease in quality<br />
    ... but generally not possible to extract a region in the
    compressed domain</p>

    <p class='phone'><cite>Raphael:</cite> it is not clear what to
    do with a spatial fragment<br />
    ... my suggestion would be that the server send the whole
    picture, but the UA does something with the fragment, e.g.
    highlight the region</p>

    <p class='phone'><cite>Davy:</cite> for a mobile use case, it
    makes more sense to not download the whole image, but just the
    region</p>

    <p class='phone'><cite>Frank:</cite> why not specifying that in
    the URI, whether the client want to download the complete
    resource or not</p>

    <p class='phone'><cite>Yves:</cite> can be done in HTTP with an
    extension<br />
    ... the discovery phase will be: server, tell me what do you
    support<br />
    ... for example, using the option method, or some parameters in
    the GET, there are many options<br />
    ... we can then implement the OPTIONS response, or a content
    negociation<br />
    ... discovery is always painful!</p>

    <p class='phone'><cite>Silvia:</cite> we always found that
    discovery was difficult<br />
    ... we had to find out which tracks were available</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; <a href=
    "http://wiki.xiph.org/index.php/ROE">http://wiki.xiph.org/index.php/ROE</a></p>

    <p class='phone'><cite>Silvia:</cite> we discussed a format,
    named ROE, which is a media file format description<br />
    ... this is currently used in Metavid<br />
    ... I'm not sure how the discovery and selection should be
    handled by URI or not<br />
    ... the UA asks for the ROE file, parse the XML and knows which
    tracks are available, the UA can then request the right
    track</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; e.g.
    ?track=a1,v1,sub1,cap1</p>

    <p class='phone'><cite>Raphael:</cite> can we find the track
    description in some headers of the container format?</p>

    <p class='phone'><cite>Davy:</cite> it depends on the format,
    it might be the case</p>

    <p class='phone'><cite>Silvia:</cite> I agree, it didn't exist
    for ogg, that's why we invented ROE<br />
    ... I'm in favor of specifying a syntax, even though just one
    format will be able to deal with it<br />
    ... so have a way of specifying tracks and we may list later on
    which codec and container formats can process</p>

    <p class='phone'><cite>Davy:</cite> the audio encoding formats
    are just relevant for the temporal dimension<br />
    ... the still images format: JPEG2000 is pretty advanced<br />
    ... the container formats: mov, mp4, 3gp allows to select track
    and names, but we need to modify some values (for example
    change the length field)<br />
    ... for other formats such as MXG, ASF, I put question
    marks</p><a name="action08" id="action08"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> Davy to complete the table, trying to
    get the answer for the current question marks, except when this
    is a close format [recorded in <a href=
    "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action08">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action08</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Created ACTION-22
    - Complete the table, trying to get the answer for the current
    question marks, except when this is a close format [on Davy Van
    Deursen - due 2008-12-16].</p>

    <p class='phone'><cite>Davy:</cite> some formats are then
    useless for our purpose, because they will support nothing
    (e.g. WAV, AIFF, AU, XMF)</p>

    <p class='phone'><cite>Raphael:</cite> it is still interested
    to report this information in the document<br />
    ... Summary: we agree to cover these 4 dimensions<br />
    ... perhaps the syntax will be simpler for the temporal
    dimension, since it will be 99% of our use cases<br />
    ... perhaps the temporal dimension will be the default
    one<br />
    ... up to decide to the WG when we will talk about the
    syntax</p>

    <p class='phone'>LUNCH TIME</p>

    <p class='phone'>Silvia, quick poll</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; yes?</p>

    <p class='phone'>Media Annotations is willing to organize the
    next joint face to face meeting in Barcelona</p>

    <p class='phone'>prior to the WWW conference in Madrid</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; awww - I'd love to
    go there!</p>

    <p class='phone'>potential dates are: 16 and 17 of April</p>

    <p class='phone'>WWW conference will then be 20-24 of April</p>

    <p class='phone'>so you have to spend the week-end in Barcelona
    and/or Madrid</p>

    <p class='phone'><cite>scribe:</cite> we can also go to the
    beach :-)</p>

    <p class='phone'>will you be able to make it ?</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; maybe</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; will need to see
    from the biz POV and whether I can get Mozilla to sponsor
    it</p>

    <p class='phone'>depends on your funding ? Mozilla?</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; (or some of it)</p>

    <p class='phone'>ok</p>

    <p class='phone'>good, but I note your interest</p>

    <h3 id="item04">4. Implementation Issues (protocol &amp;
    caching)</h3>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; when you're in
    australia, meeting people in your field is of major interest,
    since everybody is so far away</p>

    <p class='phone'>Yves leads the dicussion</p>

    <p class='phone'><cite>Yves:</cite> look at <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/wiki/HTTP_Fragment_Caches">
    http://www.w3.org/2008/WebVideo/Fragments/wiki/HTTP_Fragment_Caches</a><br />

    ... there is a discussion in the mailing list between myself,
    Silvia and others<br />
    ... about the solutions recommended by annodex, the 4-way
    handshakes<br />
    ... and I was discussing the alternative 2-way handshakes<br />
    ... both are limited, because it will be difficult to access
    track fragments, and even worst spatial fragments, because
    transcoding might be required</p>

    <p class='phone'><cite>Silvia:</cite> we should not use
    fragment when a transcoding operation is needed</p>

    <p class='phone'><cite>Yves:</cite> when you deal with tracks,
    do you think we can handle everything in the compressed
    domains?</p>

    <p class='phone'><cite>Silvia:</cite> yes, tracks are dealt
    with by the container formats</p>

    <p class='phone'><cite>Davy:</cite> yes it depends on the
    format<br />
    ... why do you think the outcome of a transcoding operation is
    not a fragment anymore ?</p>

    <p class='phone'><cite>Silvia:</cite> because there is no one
    to one mapping between the bytes of the original file and the
    outcome file<br />
    ... i'm talking about physical fragment and not logical
    fragment</p>

    <p class='phone'><cite>Yves:</cite> I argue that a fragment in
    the URI spec does not specify if it is a compressed
    resource<br />
    ... it is a part of the resource</p>

    <p class='phone'><cite>Silvia:</cite> I argue that a fragment
    of a original resource must be a part of the resource</p>

    <p class='phone'><cite>Yves:</cite> I do not argue<br />
    ... you can have lostless transformation process, but there is
    not a single byte range process</p>

    <p class='phone'><cite>Silvia:</cite> I didn't argue about
    having a single or multiple byte ranges</p>

    <p class='phone'>Yves and Davy disagree</p>

    <p class='phone'><cite>Yves:</cite> if you transcode to a
    different format, yes, this is a different resource<br />
    ... but if you transcode to the same format, I would consider
    this is the same resource<br />
    ... so a valid fragment<br />
    ... example, get all &lt;H1&gt; in a HTML page, this is a
    fragment<br />
    ... it might not be a continous fragment, so difficult to
    cache, but it is still a fragment</p>

    <p class='phone'><cite>Silvia:</cite> YES, but you're not
    changing the bytes, you have the same bytes</p>

    <p class='phone'><cite>Yves:</cite> ok, but if you use FLAC,
    which is lostless, you will have a valid fragment<br />
    ... mp3 is definitively not the same thing<br />
    ... the fact that the stored bytes are different is not
    relevant<br />
    ... the criteria is what you get, what you watch</p>

    <p class='phone'><cite>Raphael:</cite> ok, but how cache will
    handle that?</p>

    <p class='phone'><cite>Yves:</cite> caches are not forced to
    store _all_ fragments, need to be specified<br />
    ... merging does not involve a simple concatenation of bytes
    ... you already add information for serving the fragments</p>

    <p class='phone'><cite>Silvia:</cite> I think we will have a
    lot of pitfalls with this path</p>

    <p class='phone'><cite>Yves:</cite> which ones ? I want to see
    examples</p>

    <p class='phone'><cite>Silvia:</cite> take samples of a FLAC
    file, decode them, and re-encode them, you will not get a
    playable piece ???<br />
    ... I just do not see happening this extra complexity</p>

    <p class='phone'><cite>Yves:</cite> perhaps, but nothing
    prevent to do that?<br />
    ... i'm arguing that such operation done in the cache might be
    more efficient</p>

    <p class='phone'><cite>Silvia:</cite> i don't want to support
    transoding with loss of information</p>

    <p class='phone'><cite>Yves:</cite> agree, fair thing to
    do<br />
    ... but not all merging operation have to be done with
    transcoding</p>

    <p class='phone'><cite>Raphael:</cite> which formats are we
    talking about ?</p>

    <p class='phone'><cite>Davy:</cite> most of them that are used
    loose information anyway</p>

    <p class='phone'><cite>Yves:</cite> if you do a lossy
    transformation, then you will get another resource, so another
    URI<br />
    ... but it can be done transparently using content
    negotiation<br />
    ... another issue, is that annodex create an unlimited numbers
    of sub-resources<br />
    ... because it uses the ? and not a real fragment</p>

    <p class='phone'><cite>Silvia:</cite> we did that because we
    thought it was not appropriate to use the '#' ... but I'm happy
    to use now the hash</p>

    <p class='phone'><cite>Yves:</cite> i just want to come to the
    header (and footer) of the current annodex solution ... that is
    done in the compressed domain<br />
    ... the solution I'm talking about has headers modified<br />
    ... smart caches will have a way of doing merge in the
    compressed domain<br />
    ... it can be done in specialized proxies (dedicated to
    media)</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; in Ogg, it is not
    possible to have a video file with a gap at the beginning and a
    gap and a gap in the middle</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; it will not result
    in a valid resource</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; thus, if you have
    more than one segment, it needs to be done as video
    playlists</p>

    <p class='phone'><cite>Silvia:</cite> the solution we advocated
    in Annodex will not store n times the overlap</p>

    <p class='phone'>?</p>

    <p class='phone'><cite>Yves:</cite> in my solution, we will
    store just the complete playable files<br />
    ... so we are talking about the same thing, except that in my
    case, we store additional headers and footers</p>

    <p class='phone'>Hi Conrad ...</p>

    <p class='irc'>&lt;<cite>conrad</cite>&gt; hi raphael :-)</p>

    <p class='phone'><cite>Conclusion:</cite> Yves advocates to
    store and cache what the server is serving, playable resources,
    so the bytes corresponding to the fragment requested enhanced
    with the appropriate header/footer depending on the encoding
    format</p>

    <p class='phone'><cite>Yves:</cite> same as byte ranges, if
    there is overlap, the cache will merge them<br />
    ... we are talking about smart caches ... the other ones will
    not cache them</p>

    <p class='phone'><cite>Silvia:</cite> I will favor we go mainly
    for byte ranges and see immediate implementations<br />
    ... and see later what can be improved</p>

    <p class='phone'><cite>Yves:</cite> if you want to do only byte
    ranges, you should do it such a way that it is still a
    fragment<br />
    ... the solution I was advocating does that naturally, but
    requires smart caches</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; The byte range based
    proposal for caching web proxies in annodex is one that can be
    supported by existing web proxies</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; therefore I suggest
    we support that first</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; the difference
    between this and what Yves proposes is that the recomposition
    intelligence goes into the server or into the web proxy</p>

    <p class='phone'><cite>Raphael:</cite> but does annodex
    solution implies storing additional information such as
    header/footer ?</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; if Yves case, the
    web proxy has to know about all the encoding formats and needs
    to understand how to recompose them</p>

    <p class='phone'><cite>Erik:</cite> I wonder if your two
    solutions are orthogonal or not?</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; I was proposing to
    allow both</p>

    <p class='phone'><cite>Raphael:</cite> Silvia, YES, but Yves
    talked about smart dedicated web media proxies</p>

    <p class='phone'><cite>Yves:</cite> I agree with supporting
    both, but I would add we put too much emphasis on caching,
    given that most of the traffic is not cached anyway<br />
    ... so we should not spend too much time on caches</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; it's done through
    services like Akamai</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; who are not using HTTP
    caches for that (at least the CDN I know of)</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; no, they are using
    proprietary solutions</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; and thus avoiding
    the existing Web proxy infrastructure</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; but that's outside
    of what we need to worry about :)</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; yes :)</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; say hi to the media
    annotations guys :)</p>

    <p class='irc'>&lt;<cite>Silvia</cite>&gt; I will continue to
    hang out here</p>

    <p class='phone'>Coffee break</p>

    <h3 id="item05">5. Joint Session with Media Annotations WG</h3>

    <h3 id="item06">joint meeting MF &amp; MA</h3>

    <p class='irc'>&lt;<cite>erik</cite>&gt; poll third F2F:
    16/04-17/04 @ Barcelona (prior to WWW conference @ Madrid)</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; everybody finds it a
    good idea</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; Raphael to talk about
    status of MF group</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/wiki/Main_Page">http://www.w3.org/2008/WebVideo/Fragments/wiki/Main_Page</a></p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; Raphael summarizes
    wiki-pages under "Preparation of Working Draft"</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; * Use Cases</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; ... <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/wiki/Use_Cases_%26_Requirements_Draft">
    http://www.w3.org/2008/WebVideo/Fragments/wiki/Use_Cases_%26_Requirements_Draft</a></p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; ... functional and
    non-functional requirements are also part of that UC page</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; * Communication
    between client and server</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; ... <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/wiki/HTTP_implementation">
    http://www.w3.org/2008/WebVideo/Fragments/wiki/HTTP_implementation</a>
    (to be elaborated soon)</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; ... 2-way &amp; 4-way
    handshake</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; * Existing
    technologies</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; ... <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/wiki/Existing_Technologies_Survey">
    http://www.w3.org/2008/WebVideo/Fragments/wiki/Existing_Technologies_Survey</a></p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; ... cover all
    technologies out there &amp; in the end convert our solution
    back to existing ones</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; Question: what type of
    fragments will be possible?</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; ... temporal for sure
    in v1</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; ... temporal &amp;
    spatial for video is most difficult one in v2</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; ... also tracks are in
    scope in v1</p><a name="action09" id="action09"></a>

    <p class='irc'>&lt;<cite>erik</cite>&gt;
    <strong>ACTION:</strong> Erik (together with Jean-Pierre) to
    add TV-Anytime also to Existing Technologies Survey [recorded
    in <a href=
    "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action09">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action09</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Created ACTION-23
    - (together with Jean-Pierre) to add TV-Anytime also to
    Existing Technologies Survey [on Erik Mannens - due
    2008-12-16].</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; ability of using XMP
    for notition of tracks ... this seems possible (link MF &amp;
    MA) ... to be investigated</p>

    <p class='irc'>&lt;<cite>scribe</cite>&gt; scribenick: erik</p>

    <p class='phone'><cite>UNKNOWN_SPEAKER:</cite> common scenario
    from MA (description of resources) to MF (communication
    client/server through content negotiation) for selecting
    tracks</p>

    <p class='irc'>&lt;<cite>fsasaki</cite>&gt; <a href=
    "http://www.w3.org/2008/WebVideo/Annotations/wiki/XMP">http://www.w3.org/2008/WebVideo/Annotations/wiki/XMP</a></p>

    <p class='irc'>&lt;<cite>fsasaki</cite>&gt;
    Ingredients</p><a name="action10" id="action10"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> Erik (through extra info from Felix)
    to ask Adobe (Larry) more info about xmpMM:Ingredients
    [recorded in <a href=
    "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action10">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action10</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Created ACTION-24
    - (through extra info from Felix) to ask Adobe (Larry) more
    info about xmpMM:Ingredients [on Erik Mannens - due
    2008-12-16].</p>

    <p class='phone'>Felix to talk about status of MA group</p>

    <p class='phone'>* stating problem of information loss when
    mapping (setting) information from one format to generic MA
    ontology</p>

    <p class='phone'><cite>scribe:</cite> Raphael: is common subset
    on metalevel not enough?<br />
    ... felix: looked at existing meta-models today ... all
    "getting" models, not "setting" ... issues (protocol,
    information loss)<br />
    ... Raphael: maybe MF can give some input via explanation of
    our table ... within ...<br />
    ... <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/wiki/Types_of_Fragment_Addressing">
    http://www.w3.org/2008/WebVideo/Fragments/wiki/Types_of_Fragment_Addressing</a><br />

    ... summary: 5th column ... everywhere where there is a
    "1"&amp;"2" it is possible to add metadata within</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Raphael: in case
    some metadata are embedded into the header of a media resource,
    should we be able to have access to it using a fragment ?</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... using which
    dimension ? the 'name' dimension ?</p>

    <p class='phone'><cite>felix:</cite> will named fragments be
    possible?<br />
    ... Raphael: yes (i18 will be problem to handle though)</p>

    <p class='phone'>* XMP overview</p>

    <p class='phone'><cite>scribe:</cite> <a href=
    "http://www.w3.org/2008/WebVideo/Annotations/wiki/XMP">http://www.w3.org/2008/WebVideo/Annotations/wiki/XMP</a><br />

    ... Raphael: what about collisions of types/values?</p>

    <p class='irc'>&lt;<cite>Daniel</cite>&gt; <a href=
    "http://dev.w3.org/2008/video/mediaann/mediaont-api-1.0/mediaont-api-1.0.html">
    http://dev.w3.org/2008/video/mediaann/mediaont-api-1.0/mediaont-api-1.0.html</a></p>

    <p class='phone'><cite>scribe:</cite> only get-functions for
    the moment (cfr. "setting"-problem)</p>

    <p class='irc'>&lt;<cite>fsasaki</cite>&gt; <a href=
    "http://www.w3.org/2008/WebVideo/Annotations/wiki/FeaturesTable">
    http://www.w3.org/2008/WebVideo/Annotations/wiki/FeaturesTable</a></p>

    <p class='irc'>&lt;<cite>fsasaki</cite>&gt; <a href=
    "http://dev.w3.org/2008/video/mediaann/mediaont-req/mediaont-req.html">
    http://dev.w3.org/2008/video/mediaann/mediaont-req/mediaont-req.html</a></p>

    <p class='phone'><cite>scribe:</cite> UC document</p>

    <p class='phone'>formal review of MA UC Doc by MF (probably
    before 31/12/08)</p>

    <p class='phone'>formal review of MF UC Doc by MA (and others
    SVG, HTML5, TimedText) (probably before 31/01/09)</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; adjourn</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; thx the
    organizers</p>

    <p class='irc'>&lt;<cite>vmalais</cite>&gt; logout</p>
  </div>

  <h2><a name="ActionSummary" id="ActionSummary">Summary of Action
  Items</a></h2><!-- Action Items -->
  <strong>[NEW]</strong> <strong>ACTION:</strong> Davy to complete
  the table, trying to get the answer for the current question
  marks, except when this is a close format [recorded in <a href=
  "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action08">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action08</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> Erik (through
  extra info from Felix) to ask Adobe (Larry) more info about
  xmpMM:Ingredients [recorded in <a href=
  "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action10">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action10</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> Erik (together
  with Jean-Pierre) to add TV-Anytime also to Existing Technologies
  Survey [recorded in <a href=
  "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action09">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action09</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> Erik to check
  with Philippe the status of Cisco (Paul Bosso), Apple (Dave
  Singer or Eric Carlson) [recorded in <a href=
  "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action05">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action05</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> Michael to check
  with Wolfgang whether he is still interested in this WG [recorded
  in <a href=
  "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action04">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action04</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> Raphael to check
  with Karen the status of Fox Interactive, if they could have an
  interest in the group [recorded in <a href=
  "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action06">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action06</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> Raphael to see
  with Marie Claire who from Daily Motion can join [recorded in
  <a href=
  "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action02">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action02</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> Rapha�l to check
  with Karen the status of Fox Interactive, if they could have an
  interest in the group [recorded in <a href=
  "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action07">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action07</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> Troncy to check
  with Karen about Blinx joining or not W3C and Colm the WG
  [recorded in <a href=
  "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action03">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action03</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> Yves to find out
  with Philippe who from Google would be interested to join
  [recorded in <a href=
  "http://www.w3.org/2008/12/09-mediafrag-minutes.html#action01">http://www.w3.org/2008/12/09-mediafrag-minutes.html#action01</a>]<br />

  &nbsp;<br />
  [End of minutes]<br />
  <hr />

  <address>
    Minutes formatted by David Booth's <a href=
    "http://dev.w3.org/cvsweb/~checkout~/2002/scribe/scribedoc.htm">
    scribe.perl</a> version 1.133 (<a href=
    "http://dev.w3.org/cvsweb/2002/scribe/">CVS log</a>)<br />
    $Date: 2008/12/09 16:02:42 $
  </address>

  <div class="diagnostics">
    <hr />

    <h2>Scribe.perl diagnostic output</h2>[Delete this section
    before finalizing the minutes.]<br />
    <pre>
This is scribe.perl Revision: 1.133  of Date: 2008/01/18 18:48:51
Check for newer version at <a href=
"http://dev.w3.org/cvsweb/~checkout~/2002/scribe/">http://dev.w3.org/cvsweb/~checkout~/2002/scribe/</a>

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/N/?/
Succeeded: s/encoding/encapsulation/
Succeeded: s/option/OPTIONS/
Succeeded: s/will/may/
Succeeded: s/Sivlia/Silvia/
Succeeded: s/3. Implementation Issues/4. Implementation Issues/
Succeeded: s/cache/how cache/
Succeeded: s/lost/loss/
Succeeded: s/domains/domain/
Succeeded: s/anhanced/enhanced/
Succeeded: s/apropriate/sppropriate/
Succeeded: s/sppropriate/appropriate/
Succeeded: s/was/I was/
Succeeded: s/do/allow/
Found ScribeNick: raphael
Found ScribeNick: erik
Inferring Scribes: raphael, erik
Scribes: raphael, erik
ScribeNicks: raphael, erik

WARNING: Replacing list of attendees.
Old list: +0329331aaaa Tom Raphael Erik Davy Frank Yves +61.2.801.2.aabb Silvia
New list: +329331aaaa Meeting_Room +61.2.801.2.aabb Silvia

Default Present: +329331aaaa, Meeting_Room, +61.2.801.2.aabb, Silvia

WARNING: Replacing previous Present list. (Old list: Yves, Erik, Raphael, Tom, Davy, Frank_(canon_observer), Silvia, (irc))
Use 'Present+ ... ' if you meant to add people without replacing the list,
such as: &lt;dbooth&gt; Present+ Yves, Frank, Davy, Erik, Raphael, Tom, Silvia_(remote)

Present: Yves Frank Davy Erik Raphael Tom Silvia_(remote)
Agenda: <a href=
"http://www.w3.org/2008/WebVideo/Fragments/wiki/SecondF2FAgenda">http://www.w3.org/2008/WebVideo/Fragments/wiki/SecondF2FAgenda</a>
Found Date: 09 Dec 2008
Guessing minutes URL: <a href=
"http://www.w3.org/2008/12/09-mediafrag-minutes.html">http://www.w3.org/2008/12/09-mediafrag-minutes.html</a>
People with action items: davy erik extra felix from info jean-pierre l michael rapha raphael through together troncy with yves

</pre>[End of <a href=
"http://dev.w3.org/cvsweb/~checkout~/2002/scribe/scribedoc.htm">
    scribe.perl</a> diagnostic output]
  </div>
</body>
</html>