17-mediafrag-minutes.html 44.6 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 -- 17 Apr
  2009</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>Media Fragments Working Group Teleconference</h1>

  <h2>17 Apr 2009</h2>

  <p><a href=
  'http://www.w3.org/2008/WebVideo/Fragments/wiki/ThirdF2FAgenda'>Agenda</a></p>

  <p>See also: <a href=
  "http://www.w3.org/2009/04/17-mediafrag-irc">IRC log</a></p>

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

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

      <dd>Erik, Davy, Guillaume, Michael, Jack, Conrad, Raphael,
      Frank_(observer), Yves_(remote), Silvia_(remote), Dave</dd>

      <dt>Regrets</dt>

      <dt>Chair</dt>

      <dd>Erik, Raphael</dd>

      <dt>Scribe</dt>

      <dd>Raphael</dd>
    </dl>
  </div>

  <h2>Contents</h2>

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

      <ol>
        <li><a href="#item01">1. Quick summary of First day
        meeting</a></li>

        <li><a href="#item02">2. Implementation Reports</a></li>

        <li><a href="#item03">Test Cases</a></li>

        <li><a href="#item04">4. Issues</a></li>

        <li><a href="#item05">5. Wrap Up</a></li>

        <li><a href="#item06">AOB</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: 17 April
    2009</p>

    <p class='phone'>Silvia, you miss some amazing food yesterday
    night, but you will see the pretty pictures taken :-)</p>

    <p class='phone'>you miss as well the longggggggggggggg drink
    session :-)</p>

    <p class='phone'>trackbot, start telecon</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Meeting: Media
    Fragments Working Group Teleconference</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Date: 17 April
    2009</p>

    <p class='irc'>&lt;<cite>silvia</cite>&gt; any changes to the
    agenda?</p>

    <p class='irc'>&lt;<cite>scribe</cite>&gt; Meeting: Media
    Fragments WG, 3rd F2F Meeting - Barcelona (Spain)</p>

    <p class='phone'>yes Silvia, we will start with the session:
    Implementation Reports</p>

    <p class='phone'><cite>scribe:</cite> then talk about the
    issues<br />
    ... and finally about the test cases, where we need to kick off
    the work</p>

    <p class='phone'>Does that suit you ?</p>

    <p class='irc'>&lt;<cite>silvia</cite>&gt; sure</p>

    <p class='phone'>Silvia, we are waiting for Jack, coming in a
    few minutes</p>

    <p class='phone'><cite>scribe:</cite> and Yves, that he could
    phone us</p>

    <p class='irc'>&lt;<cite>scribe</cite>&gt; Scribe: Raphael</p>

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

    <h3 id="item01">1. Quick summary of First day meeting</h3>

    <p class='phone'><cite>Raphael:</cite> we should have a picture
    representing the general structure of the media object, the H,
    H', K, D, etc. we talked about yesterday<br />
    ... Can we do such a representation valid for most of the
    formats?</p>

    <p class='phone'><cite>Conrad:</cite> yes, for the streamable
    formats<br />
    ... I'm happy to take an action to draw such a schema</p>

    <p class='phone'>trackbot, status?</p><a name="action01" id=
    "action01"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> Conrad to draw a representation of the
    general structure of a media resource, for streamable formats
    [recorded in <a href=
    "http://www.w3.org/2009/04/17-mediafrag-minutes.html#action01">http://www.w3.org/2009/04/17-mediafrag-minutes.html#action01</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Created ACTION-69
    - Draw a representation of the general structure of a media
    resource, for streamable formats [on Conrad Parker - due
    2009-04-24].</p>

    <h3 id="item02">2. Implementation Reports</h3>

    <p class='phone'>a) Davy wants to present and demo a software
    that allows to slide media files independently of the codec
    format</p>

    <p class='phone'>b) Jack has some reporting to do on its
    implemntation</p>

    <p class='phone'>c) Davy wants to talk about RTSP
    implementation</p>

    <p class='phone'>d) Conrad wants to report on his experience in
    Annodex</p>

    <p class='phone'>Silvia, Davy will start with some slides that
    are coming</p>

    <p class='phone'>Davy going through a presentation</p>

    <p class='phone'>Slides at: <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/meetings/2009-04-16-f2f_barcelona/Adaptation_logic_for_(server-side)_media_fragment_extraction.pdf">
    http://www.w3.org/2008/WebVideo/Fragments/meetings/2009-04-16-f2f_barcelona/Adaptation_logic_for_(server-side)_media_fragment_extraction.pdf</a></p>

    <p class='phone'><cite>Davy:</cite> Slide 6, my PhD work is
    built on BSD content adaptation, using RDF<br />
    ... based on a model for media bitstreams encoded in OWL<br />
    ... Slide 7: Coding formats -&gt; model for media bitstreams
    -&gt; container formats<br />
    ... Slide 8: model for media bitsreams represented as a set of
    classes and properties<br />
    ... left part is the structure part, while the right part is
    more semantic<br />
    ... Slide 9: shows the workflow<br />
    ... Slide 10: shows an example, two tracks (audio, video)
    represented in the intermediary model<br />
    ... assume there is a mapping between one particular codec
    format and this model<br />
    ... Track selection, means selection of MediaBitstreams,
    straightforward<br />
    ... Temporal selection, means selection of data blocks, based
    on their timestamps<br />
    ... problem for Spatial clipping?</p>

    <p class='phone'><cite>Jack:</cite> it works only for scalable
    formats for Spatial Fragment</p>

    <p class='phone'><cite>Davy:</cite> Slide 11: general
    architecture of the NinSuna platform<br />
    ... the retrieval service is not yet compatible with the media
    fragment syntax, but we could adapt it<br />
    ... Slide 12: two remarks, 1/ is the media needs to be
    'ingested', i.e. the intermediary model needs to be
    instantiated and 2/ currently client decides the output
    format</p>

    <p class='phone'>Time for demo</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; DON'T CLICK on
    the following link, now:</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; <a href=
    "http://respighi.elis.ugent.be/NinSunaWeb">http://respighi.elis.ugent.be/NinSunaWeb</a></p>

    <p class='irc'>&lt;<cite>silvia</cite>&gt; what is the link
    for?</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; silvia: that's
    davy's demo :)</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; note that the
    demo requires IE and Silverlight</p>

    <p class='irc'>&lt;<cite>silvia</cite>&gt; ah, thanks</p>

    <p class='phone'><cite>scribe:</cite> and VLC for the
    playback</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; raphael, no</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; we will put thata on
    another server</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; our demo can be found
    at <a href=
    "http://multimedialab.elis.ugent.be/NinSuna/">http://multimedialab.elis.ugent.be/NinSuna/</a></p>

    <p class='irc'>&lt;<cite>silvia</cite>&gt; is there a
    screencast?</p>

    <p class='irc'>&lt;<cite>silvia</cite>&gt; erik?</p>

    <p class='phone'>A screencast will be made by the IBBT guys
    next week</p>

    <p class='irc'>&lt;<cite>silvia</cite>&gt; ok - those
    screenshots at the demo site are good, too</p>

    <p class='phone'>I'm uploading the videos on my web page</p>

    <p class='phone'>cofffee break, we are back in 15 min</p>

    <p class='phone'>Videos will be at <a href=
    "http://homepages.cwi.nl/~troncy/tmp/spool/video/">http://homepages.cwi.nl/~troncy/tmp/spool/video/</a></p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt;
    coffeeeeeeeeeeee</p>

    <p class='phone'>Silvia, we will resume</p>

    <p class='phone'><cite>Raphael:</cite> now we will listen to
    Jack</p>

    <p class='phone'><cite>Jack:</cite> I took the Python URL
    library<br />
    ... I took this library and add some code of myself<br />
    ... for parsing the media fragments part<br />
    ... and here I discover all the issues discovered and reported,
    now fixed<br />
    ... and then the problems started, I tried first with the file
    URL (local resource)<br />
    ... problem is with synch on video/audio, should we do
    transcoding or not, etc.<br />
    ... my goal was to first address:
    file://myvideo.mp4#t=10,20<br />
    ... use ffmpeg for slicing the media<br />
    ... it's hard to use it, the API changes frequently, but the
    functionalities are very good<br />
    ... at the C level, it does what we want, but at the command
    level, it is meant to do transcoding</p>

    <p class='phone'><cite>Guillaume:</cite> why not using the
    Python library G-streams, an API for ffmpeg?</p>

    <p class='phone'>URL lib for Python: <a href=
    "http://docs.python.org/library/urllib.html">http://docs.python.org/library/urllib.html</a></p>

    <p class='irc'>&lt;<cite>guillaume</cite>&gt; Gstreamer Python
    binding library : <a href=
    "http://gstreamer.freedesktop.org/modules/gst-python.html">http://gstreamer.freedesktop.org/modules/gst-python.html</a></p>

    <p class='irc'>&lt;<cite>guillaume</cite>&gt; and GStreamer
    itself <a href=
    "http://www.gstreamer.net/">http://www.gstreamer.net/</a></p>

    <p class='phone'><cite>Jack:</cite> I will have a working
    version in a few more days, but it will transcode<br />
    ... since this is what ffmepg does<br />
    ... my plan is thus rather going through C++ implementation, to
    attack ffmeph low layers, and have no transcoding<br />
    ... I can share the code that does the parsing of the
    fragment</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; Anyone done or
    plans to do a client-side implementation of the parsing, etc.,
    say in JavaScript</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; if not, I might
    have some resources (not myself, for sure, but a PhD of mine)
    who could do something based on jQuery</p><a name="action02"
    id="action02"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> Jack to commit in CVS (code directory)
    his python code doing the parsing on client side of the media
    fragment [recorded in <a href=
    "http://www.w3.org/2009/04/17-mediafrag-minutes.html#action02">http://www.w3.org/2009/04/17-mediafrag-minutes.html#action02</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Created ACTION-70
    - Commit in CVS (code directory) his python code doing the
    parsing on client side of the media fragment [on Jack Jansen -
    due 2009-04-24].</p><a name="action03" id="action03"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> Michael to investigate whether he
    could have an implementation in Javascript that does the
    client-side media fragments implementation [recorded in
    <a href="http://www.w3.org/2009/04/17-mediafrag-minutes.html#action03">
    http://www.w3.org/2009/04/17-mediafrag-minutes.html#action03</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Created ACTION-71
    - Investigate whether he could have an implementation in
    Javascript that does the client-side media fragments
    implementation [on Michael Hausenblas - due 2009-04-24].</p>

    <p class='phone'>Conrad will now report on his implementation
    experience</p>

    <p class='phone'>Davy, first report on RTSP implementation</p>

    <p class='phone'><a href=
    "http://homepages.cwi.nl/~troncy/Talks/2009-03-06-mozcamp/#(3)">
    http://homepages.cwi.nl/~troncy/Talks/2009-03-06-mozcamp/#(3)</a></p>

    <p class='phone'><cite>Davy:</cite> RTSP implementation wiki
    page, <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/wiki/RTSP_implementation">
    http://www.w3.org/2008/WebVideo/Fragments/wiki/RTSP_implementation</a><br />

    ... quick test, I have an implementation of the media fragment
    over the RTSP protocol<br />
    ... it works for track and temporal selection<br />
    ... first communication is the Describe, and the server answers
    which track is available, etc.<br />
    ... then for each track, we have a Setup<br />
    ... so we can have easily a track selection in RTSP<br />
    ... next command is the play command, where we can specify the
    temporal clipping<br />
    ... I don't remember which units is supported, but for sure npt
    and smpte<br />
    ... again, I have an implementation that parses the media
    fragment, and generates a PLAY command<br />
    ... i do not yet cover the whole grammar of the media fragment,
    so far just, t= and track=<br />
    ... implemented in C<br />
    ... I'm working on covering the whole grammar</p>

    <p class='phone'><cite>Raphael:</cite> my (silly?) idea, embed
    VLC in a web page<br />
    ... implement on client side, in the browser, the possibility
    to write down a media fragment URI (rtsp), use the code of Davy
    to generate the good Describe, Setup and Play commands<br />
    ... send that through VLC and get the results back, using the
    VLC plugin in the browser</p>

    <p class='phone'><cite>Davy:</cite> should be doable<br />
    ... my issue is with track selection, how do you know the name
    of the track</p>

    <p class='phone'><cite>Jack:</cite> this is related to the
    issue raised by Silvia, <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/tracker/issues/4">http://www.w3.org/2008/WebVideo/Fragments/tracker/issues/4</a><br />

    ... what I come accoss, is that often, you want not 1 track,
    but 2 tracks, or all tracks except one<br />
    ... not very convenient with our current scheme?</p>

    <p class='phone'><cite>Silvia:</cite> suggestion to add a comma
    separation, such as: #track='track1,track2,track3'</p>

    <p class='phone'><cite>Jack:</cite> or event having a minus</p>

    <p class='irc'>&lt;<cite>silvia</cite>&gt; the naming of the
    tracks is up to the container format - some allow it, others
    don't</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; Scribenick:
    mhausenblas</p>

    <p class='irc'>&lt;<cite>silvia</cite>&gt; for example
    QuickTime has chapter tracks that should be addressable in the
    given way with #track</p>

    <p class='phone'>;)</p>

    <p class='phone'>Conrad has essentially said everything
    yesterday, now no new stuff</p>

    <p class='irc'>&lt;<cite>silvia</cite>&gt; Jack: mentions that
    naming of tracks is decided by author and they create the
    URL</p>

    <p class='irc'>&lt;<cite>silvia</cite>&gt; Conrad: mentions
    that with Accept-Language it should be possible to leave the
    track selection for languages to the server</p>

    <p class='phone'><cite>raphael:</cite> conrad ok for you to
    implement it based on # rather than ?</p>

    <p class='phone'><cite>conrad:</cite> will try to implement the
    hash part as well</p>

    <p class='phone'><cite>raphael:</cite> agree, after the first
    step both # and ? have the same procedure<br />
    ... conrad would you implement the 'missing link' using the
    HTTP Link: header draft</p>

    <p class='phone'><cite>conrad:</cite> yes, I like Yves idea
    using it, gonna try</p>

    <p class='phone'><cite>raphael:</cite> Yves, silvia any
    question?</p>

    <h3 id="item03">Test Cases</h3>

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

    <p class='phone'><cite>yves:</cite> give a set of URI's and
    describe what should be expected</p>

    <p class='phone'><cite>michael:</cite> rather have a set of
    simple cases to start with</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; parsing URI is more or
    less done (via the code I contributed), so it's more how a
    cient parse que URI and act based on that</p>

    <p class='phone'><cite>michael:</cite> really stress on what
    the outcome is<br />
    ... try to make it as automatic as possible</p>

    <p class='phone'><cite>Jack:</cite> some temporal testing
    should at least be able to be done automatically (cfr. via
    using subtitles)<br />
    ... some manual video-inspection still has to be done</p>

    <p class='irc'>&lt;<cite>conrad</cite>&gt; we could add time
    range validation to <a href=
    "http://validator.annodex.org/">http://validator.annodex.org/</a>
    :-)</p>

    <p class='phone'><cite>Jack:</cite> because we use half-open
    intervals</p>

    <p class='irc'>&lt;<cite>conrad</cite>&gt; <a href=
    "http://validator.annodex.org/?uri=http%3A%2F%2Fia331343.us.archive.org%2F2%2Fitems%2Fnight_of_the_living_dead%2Fnight_of_the_living_dead.ogv%3Ft%3D10%2C20">
    http://validator.annodex.org/?uri=http%3A%2F%2Fia331343.us.archive.org%2F2%2Fitems%2Fnight_of_the_living_dead%2Fnight_of_the_living_dead.ogv%3Ft%3D10%2C20</a></p>

    <p class='phone'>michael (on blackboard)</p>

    <p class='phone'>on UA:</p>

    <p class='irc'>&lt;<cite>jackjansen</cite>&gt; erik, I wanted
    to say "because we use half-open intervals we can also do some
    automatic testing on video/audio"</p>

    <p class='phone'>two test cases: URI resolving &amp; Generate
    HTTP request (also see photo)</p>

    <p class='phone'>on Server:</p>

    <p class='phone'>from HTTP request to HTTP response</p>

    <p class='phone'>back on UA:</p>

    <p class='phone'>comment from jack: this is testing from User
    Expectation ... shouldn't we just test our Spec?</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; we need to test if
    it's feasible to get interoperable implementations</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; so we test the spec,
    but also (parts of) implementations as well</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; we won't test that
    implementations are 100% correct</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; we just test 100%
    coverage of the spec by interoperable implementations</p>

    <p class='phone'><cite>michael:</cite> try the top-down
    approach (opposed to Jack's proposal of doing it bottom-up)</p>

    <p class='phone'><cite>raphael:</cite> in the end both should
    come up with the same results though</p>

    <p class='phone'>(human in the loop) on UA</p>

    <p class='irc'>&lt;<cite>conrad</cite>&gt; on the last step
    ("human in the loop") we should document the expected behaviour
    of a UA: that the first frame rendered is the first frame of
    the interval, that the video image is not broken (ie. the
    decoder has been primed with the previous i-frame)</p>

    <p class='phone'>TestCase 0: URI#</p>

    <p class='phone'>-&gt; == URI == entire resource</p>

    <p class='phone'><cite>Jack:</cite> at very least for every
    dimension we should define all the test cases where we get back
    the whole resource &amp; also the ones where we get back an
    empty resource</p>

    <p class='irc'>&lt;<cite>guillaume</cite>&gt; see <a href=
    "http://www.w3.org/2006/07/SWD/RDFa/testsuite/">http://www.w3.org/2006/07/SWD/RDFa/testsuite/</a>
    for example</p>

    <p class='irc'>&lt;<cite>guillaume</cite>&gt; <a href=
    "http://torrez.us/code/rdfa-js/tests/">http://torrez.us/code/rdfa-js/tests/</a>
    and <a href=
    "http://rdfa.digitalbazaar.com/rdfa-test-harness/">http://rdfa.digitalbazaar.com/rdfa-test-harness/</a></p>

    <p class='phone'><cite>Yves:</cite> there are tools online that
    check HTTP responses</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; I agree we just need
    to test that the headers we cared for are ok</p>

    <p class='phone'><cite>michael:</cite> we should test our extra
    HTTP-headers for example ... not the complete HTTP stack</p>

    <p class='phone'><cite>raphael:</cite> practically ... just
    test specific HTTP-headers with name &amp; their possible
    values</p>

    <p class='phone'>Jack &amp; Conrad: tests are most easily done
    on server</p>

    <p class='irc'>&lt;<cite>guillaume</cite>&gt; see <a href=
    "http://wiki.csswg.org/test/harness">http://wiki.csswg.org/test/harness</a></p>

    <p class='phone'><cite>conrad:</cite> two test harnesses ...
    one for client &amp; one for server</p>

    <p class='phone'>conrad &amp; jack: both test harnesses can be
    scripted</p>

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

    <p class='irc'>&lt;<cite>jackjansen</cite>&gt; I can be
    scripted, but the qeustion is: do I do what you
    expect.....</p><a name="action04" id="action04"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> michael to setup template for test
    cases on wiki [recorded in <a href=
    "http://www.w3.org/2009/04/17-mediafrag-minutes.html#action04">http://www.w3.org/2009/04/17-mediafrag-minutes.html#action04</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Created ACTION-72
    - Setup template for test cases on wiki [on Michael Hausenblas
    - due 2009-04-24].</p>

    <p class='phone'><cite>raphael:</cite> initial 10 test cases
    ... test case 0 &amp; empty/entire resource &amp; only temporal
    and npt-units</p>

    <p class='irc'>&lt;<cite>conrad</cite>&gt; if we are testing a
    client, the base uri is the uri of the harness</p>

    <p class='irc'>&lt;<cite>conrad</cite>&gt; if we are testing a
    server, we need to provide media and test fragment requests to
    that media</p>

    <p class='phone'><cite>jack:</cite> why not only within test
    harness just the fragment (not the whole URI) to make it
    automatically testable</p>

    <p class='phone'><cite>erik:</cite> what about RTSP?</p>

    <p class='phone'><cite>jack:</cite> we do end-to-end test for
    multiple protocols</p>

    <p class='phone'><cite>raphael:</cite> how was the test suite
    handled within SMIL?</p>

    <p class='phone'><cite>Jack:</cite> SMIL 2.0 test suite was xml
    &amp; then parsed by bits of perl/python<br />
    ... then put into HTML forms, again some script processing
    &amp; afterwards put into a document ... all testing was done
    manually :(<br />
    ... please, choose the right media to run the tests!</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; invite zakim
    #mediafrag</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; trackbot, start
    telecon</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Meeting: Media
    Fragments Working Group Teleconference</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Date: 17 April
    2009</p>

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

    <p class='phone'>Wrapping up the test cases</p>

    <p class='phone'><cite>Conrad:</cite> have a test case with a
    #foo fragment<br />
    ... on a non-media resource, if the behaviour is as
    expected</p>

    <p class='phone'><cite>Dave:</cite> test whether the
    video.mp4#t=10 (as used by Google video) still works as
    expected</p>

    <p class='phone'><cite>Michael:</cite> we should completement
    with more negative test cases</p>

    <h3 id="item04">4. Issues</h3>

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

    <p class='phone'>Issue 3: Michael, <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/tracker/issues/3">http://www.w3.org/2008/WebVideo/Fragments/tracker/issues/3</a></p>

    <p class='phone'><cite>Michael:</cite> in RFC3986, the
    semantics of the fragment is per design not defined<br />
    ... together with the mime-type, one can learn what the
    semantics means<br />
    ... example, HTML, with its registration<br />
    ... but this is not true for most of the media types<br />
    ... to register at IANA and IETF the media fragment syntax</p>

    <p class='phone'><cite>Jack:</cite> this is not possible, mp4
    has already its post-hash syntax<br />
    ... we could write a cover letter, explaining who we are, and
    recommend to the mime-type owners what they should register</p>

    <p class='phone'><cite>Michael:</cite> there are plenty, is
    this possible at all?<br />
    ... it will take too much time</p>

    <p class='phone'>Member only link: <a href=
    "http://www.w3.org/2009/02/06-swcg-minutes.html#item03">http://www.w3.org/2009/02/06-swcg-minutes.html#item03</a></p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; proposal:
    media/fragment</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt;
    media/fragment+video</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt;
    media/fragment+audio</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt;
    media/fragment+image</p>

    <p class='phone'><cite>Conrad:</cite> it's a hack to use the
    Accept header for that<br />
    ... we need to do a media specific documentation</p>

    <p class='phone'><cite>Davy:</cite> problem, MPEG-21 has done
    that for MPEG resources, why they would change and adopt our
    scheme?</p>

    <p class='phone'><cite>Conrad:</cite> we have no authority, we
    could just convince them with existing implementations, and
    deployment</p>

    <p class='phone'><cite>Michael:</cite> I need to take my action
    seriously and come back with figures<br />
    ... how many mime type owners we need to talk with ?<br />
    ... how many mime-types will be affected?</p>

    <p class='phone'><cite>Jack:</cite> looking at IANA database,
    it is very hard to guess who is the owner</p>

    <p class='phone'><cite>Michael:</cite> my method, look at all
    the mime-type concerned (audio/*, video/*, image/*) and look
    for each whether they have a semantics for the hash<br />
    ... if NOT, no problem<br />
    ... if YES, then, identify the owner and evaluate the
    impact<br />
    ... evaluate impact meaning detecting clash</p>

    <p class='phone'><cite>Conrad:</cite> we should talk to the AVT
    group in IETF<br />
    ... charter: <a href=
    "http://www.ietf.org/html.charters/avt-charter.html">http://www.ietf.org/html.charters/avt-charter.html</a><br />

    ... group home page: <a href=
    "http://tools.ietf.org/wg/avt/">http://tools.ietf.org/wg/avt/</a><br />

    ... I think Dave Singer is somehow associated to this group</p>

    <p class='phone'><cite>Dave:</cite> can you confirm the
    asumption of Conrad just above?</p>

    <p class='phone'>Issue 4: <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/tracker/issues/4">http://www.w3.org/2008/WebVideo/Fragments/tracker/issues/4</a></p>

    <p class='irc'>&lt;<cite>dsinger</cite>&gt; I am looking for
    the assumption</p>

    <p class='phone'>Are you associated to some extent to the AVT
    group in IETF ?</p>

    <p class='phone'><cite>Raphael:</cite> I feel the Issue 4 is
    mixing two things</p>

    <p class='phone'>a) the fact of selecting various tracks, or
    all tracks except 1</p>

    <p class='phone'>b) pre-selecting track names</p>

    <p class='phone'><cite>Jack:</cite> for a) I think it should go
    to the version 2 of Media Fragments<br />
    ... because we will end up soon to a presentation language
    behind the hash<br />
    ... the problem is we cannot measure what amount of work it
    represents, nor to the people we need to talk too</p>

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

    <p class='phone'><cite>Davy:</cite> why not doing like RTSP
    ...</p>

    <p class='phone'><cite>Raphael:</cite> this is the ROE way</p>

    <p class='phone'><cite>Jack:</cite> we assume there is a
    mechanism for which the client can have a description of the
    media items</p>

    <p class='phone'><cite>Conrad:</cite> clarification, the
    retrieval of the description is not in the media fragments, it
    is done before</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; <a href=
    "http://www.iana.org/assignments/media-types/video/">http://www.iana.org/assignments/media-types/video/</a></p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; so it includes lots of
    RFCs (not easy to republish)</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; Michael: re
    issue 3, I mean it is always good to write the cover letter</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; ... just a
    matter of how the odds are that we can convince them, looking
    at the sheer number of potential clashes</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; ... where a
    clash is defined as a media type that defines frag
    semantics</p>

    <p class='phone'>Yves, hopefully, there will be 0 or 1 RFC
    impacted</p>

    <p class='phone'><cite>scribe:</cite> we really care when there
    is a clash</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; ... and these
    semantics contradict with ours</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; indeed</p>

    <p class='phone'><cite>Jack:</cite> no, the syntax will not
    change<br />
    ... we pass verbatim the track="value" to the server, and this
    is up to the server, capabilities of container formats</p>

    <p class='phone'><cite>Conrad:</cite> I agree with you<br />
    ... I think what we can get is just no audio or no video</p>

    <p class='phone'><cite>Jack:</cite> I dont know what get
    'audio' means in the general case</p><a name="action05" id=
    "action05"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> Conrad to change the phrasing of the
    issue 4 (just audio/video)? [recorded in <a href=
    "http://www.w3.org/2009/04/17-mediafrag-minutes.html#action05">http://www.w3.org/2009/04/17-mediafrag-minutes.html#action05</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Created ACTION-73
    - Change the phrasing of the issue 4 (just audio/video)? [on
    Conrad Parker - due 2009-04-24].</p>

    <p class='phone'>Issue 6: <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/tracker/issues/6">http://www.w3.org/2008/WebVideo/Fragments/tracker/issues/6</a></p>

    <p class='phone'><cite>Jack:</cite> the issue comes during the
    implementation trial, where I happily do some transcoding</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; my point was... "it is
    fuzzy"</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; +q to ask for
    sanity check</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; I can point to...
    <a href=
    "http://tools.ietf.org/html/draft-ietf-httpbis-p6-cache-06#section-3.6">
    http://tools.ietf.org/html/draft-ietf-httpbis-p6-cache-06#section-3.6</a></p>

    <p class='phone'><cite>Raphael:</cite> yes, thus the action-62,
    <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/tracker/actions/62">http://www.w3.org/2008/WebVideo/Fragments/tracker/actions/62</a>
    now associated</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; Warning: 214
    Transformation applied</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; (note that is is also
    in rfc2616)</p>

    <p class='irc'>&lt;<cite>conrad</cite>&gt; 4.3.2 Receipt of
    Warning: 214 Transformation Applied</p>

    <p class='irc'>&lt;<cite>conrad</cite>&gt; If the response
    includes a Warning: 214 Transformation Applied HTTP header,
    proxies must not apply further transformation.</p>

    <p class='phone'>Proposal from Yves: not really implemented,
    but we could use the warning header to warn the UA there has
    been some transcoding</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; oh proxies can do
    further transformations</p>

    <p class='irc'>&lt;<cite>Zakim</cite>&gt; mhausenblas, you
    wanted to ask for sanity check</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; it just warns the
    client that there was a transformation somewhere in the
    chain</p>

    <p class='irc'>&lt;<cite>conrad</cite>&gt; i was quoting from
    <a href=
    "http://www.w3.org/TR/ct-guidelines/">http://www.w3.org/TR/ct-guidelines/</a></p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; Michael: is
    this within our scope?</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; ... would like
    to see WG saying: YES!</p>

    <p class='phone'><cite>Jack:</cite> should we care, as a
    standardisation group, whether there is transcoding or not</p>

    <p class='irc'>&lt;<cite>conrad</cite>&gt; for formats that do
    not encode presentation in-points, we could add an HTTP
    response header to inform the client of the presentation time
    at which to enable rendering</p>

    <p class='phone'>Issue 5: <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/tracker/issues/5">http://www.w3.org/2008/WebVideo/Fragments/tracker/issues/5</a></p>

    <p class='phone'><cite>Jack:</cite> this is just a heads up,
    for us<br />
    ... we must remember if the server has done some clipping, it
    must send also the original w/h to the client<br />
    ... in the case where the spatial cropping has been asked in
    terms of %</p>

    <p class='phone'><cite>Raphael:</cite> my concern, are we
    stretching too much the definition of a fragment, when the
    fragment (= spatial clipping) has no single bytes in common
    from the original resource<br />
    ... I hope this will be answered by Yve's action to TAG<br />
    ... perhaps, we will say, this is not anymore a fragment, and
    recommend to use the ? rather than the # for this use case</p>

    <p class='phone'><cite>Frank:</cite> this is an analogy of what
    is said for spatial but not said for temporal</p>

    <h3 id="item05">5. Wrap Up</h3>

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

    <p class='phone'>raphael on the blackboard</p>

    <p class='phone'><cite>raphael:</cite> uc &amp; req doc will be
    split<br />
    ... use cases and requirements and side conditions will be in a
    separate document<br />
    ... technologies survey in another document</p>

    <p class='phone'><cite>guillaume:</cite> uc &amp; req should be
    first, then side conditions</p>

    <p class='phone'><cite>raphael:</cite> historically, side
    conditions were first</p>

    <p class='phone'><cite>mhausenblas:</cite> put survey in the uc
    &amp; req doc</p>

    <p class='phone'><cite>proposal:</cite> uc &amp; req doc and
    the technologies survey in one doc?</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; +1</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; +1</p>

    <p class='phone'>+1</p>

    <p class='phone'><cite>guillaume:</cite> duplicate side
    conditions (both in working note and REC docs)</p>

    <p class='irc'>&lt;<cite>jackjansen</cite>&gt; +1</p>

    <p class='phone'><cite>jackjansen:</cite> in the rec, side
    conditions will be specified more formally</p>

    <p class='phone'><cite>raphael:</cite> spec will be rec</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; wg note as well for
    UC&amp;req&amp;survey</p>

    <p class='phone'><cite>raphael:</cite> uc &amp; req will be
    working note?</p>

    <p class='phone'><cite>all:</cite> working note</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; should <a href=
    "http://www.w3.org/2005/Incubator/app-backplane/">http://www.w3.org/2005/Incubator/app-backplane/</a>
    be in</p><a name="action06" id="action06"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> raphael and erik to request feedback
    of other groups such as SYMM, SVG, HTML + WHATWG, WAI, MAWG,
    TAG, MobileWeb, TimedText, ... [recorded in <a href=
    "http://www.w3.org/2009/04/17-mediafrag-minutes.html#action06">http://www.w3.org/2009/04/17-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>scribe</cite>&gt;
    <strong>ACTION:</strong> Erik and Raphael to request feedback
    of other groups such as SYMM, SVG, HTML + WHATWG, WAI, MAWG,
    TAG, MobileWeb, TimedText [recorded in <a href=
    "http://www.w3.org/2009/04/17-mediafrag-minutes.html#action07">http://www.w3.org/2009/04/17-mediafrag-minutes.html#action07</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Created ACTION-74
    - And Raphael to request feedback of other groups such as SYMM,
    SVG, HTML + WHATWG, WAI, MAWG, TAG, MobileWeb, TimedText [on
    Erik Mannens - due 2009-04-24].</p>

    <p class='phone'><cite>raphael:</cite> planning for the next 6
    months<br />
    ... second public wd within 2-3 months</p>

    <p class='phone'><cite>jackjansen:</cite> 2nd wd in
    september</p>

    <p class='phone'><cite>raphael:</cite> if people do not provide
    feedback, we should find out why<br />
    ... Goal is to have a 2nd WD early in september</p>

    <p class='phone'><cite>erik:</cite> will the document already
    be split?</p>

    <p class='phone'><cite>raphael:</cite> yes, what about test
    cases?</p>

    <p class='phone'><cite>mhausenblas:</cite> would leave it in
    the wiki for a while</p>

    <p class='phone'><cite>raphael:</cite> current policy regarding
    editorship<br />
    ... which policy should we have?</p>

    <p class='phone'><cite>jackjansen:</cite> list everyone in the
    group or list everyone contributed a chapter</p>

    <p class='phone'><cite>mhausenblas:</cite> if someone writes
    something in the rec, then he/she is an editor<br />
    ... have two editors, then the contributers</p>

    <p class='phone'><cite>jackjansen:</cite> then we can assume
    that everyone is a contributor</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; yes, all the WG
    members at least</p>

    <p class='phone'><cite>raphael:</cite> looking at other w3c
    docs, they use editors and a number of contributors</p>

    <h3 id="item06">AOB</h3>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Raphael: proposal
    on the blackboard is to have a number of editors (chairs ? team
    contact? others?) and all group members as
    contributors</p><a name="action08" id="action08"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> raphael to split the document and
    setup the diff feature of xmlspec [recorded in <a href=
    "http://www.w3.org/2009/04/17-mediafrag-minutes.html#action08">http://www.w3.org/2009/04/17-mediafrag-minutes.html#action08</a>]</p>

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

    <p class='irc'>&lt;<cite>raphael</cite>&gt; trackbot,
    status?</p>

    <p class='irc'>&lt;<cite>mhausenblas</cite>&gt; <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/wiki/Semantics">http://www.w3.org/2008/WebVideo/Fragments/wiki/Semantics</a></p><a name="action09"
    id="action09"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> raphaël to split the document and
    setup the diff feature of xmlspec [recorded in <a href=
    "http://www.w3.org/2009/04/17-mediafrag-minutes.html#action09">http://www.w3.org/2009/04/17-mediafrag-minutes.html#action09</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Created ACTION-75
    - Split the document and setup the diff feature of xmlspec [on
    Raphaël Troncy - due 2009-04-24].</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; [adjourned]</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; BIG THANKS for the
    local organizers</p>

  </div>

  <h2><a name="ActionSummary" id="ActionSummary">Summary of Action
  Items</a></h2><!-- Action Items -->
  <strong>[NEW]</strong> <strong>ACTION:</strong> Conrad to change
  the phrasing of the issue 4 (just audio/video)? [recorded in
  <a href=
  "http://www.w3.org/2009/04/17-mediafrag-minutes.html#action05">http://www.w3.org/2009/04/17-mediafrag-minutes.html#action05</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> Conrad to draw a
  representation of the general structure of a media resource, for
  streamable formats [recorded in <a href=
  "http://www.w3.org/2009/04/17-mediafrag-minutes.html#action01">http://www.w3.org/2009/04/17-mediafrag-minutes.html#action01</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> Erik and Raphael
  to request feedback of other groups such as SYMM, SVG, HTML +
  WHATWG, WAI, MAWG, TAG, MobileWeb, TimedText [recorded in
  <a href="http://www.w3.org/2009/04/17-mediafrag-minutes.html#action07">
  http://www.w3.org/2009/04/17-mediafrag-minutes.html#action07</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> Jack to commit in
  CVS (code directory) his python code doing the parsing on client
  side of the media fragment [recorded in <a href=
  "http://www.w3.org/2009/04/17-mediafrag-minutes.html#action02">http://www.w3.org/2009/04/17-mediafrag-minutes.html#action02</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> Michael to
  investigate whether he could have an implementation in Javascript
  that does the client-side media fragments implementation
  [recorded in <a href=
  "http://www.w3.org/2009/04/17-mediafrag-minutes.html#action03">http://www.w3.org/2009/04/17-mediafrag-minutes.html#action03</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> michael to setup
  template for test cases on wiki [recorded in <a href=
  "http://www.w3.org/2009/04/17-mediafrag-minutes.html#action04">http://www.w3.org/2009/04/17-mediafrag-minutes.html#action04</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> raphael and erik
  to request feedback of other groups such as SYMM, SVG, HTML +
  WHATWG, WAI, MAWG, TAG, MobileWeb, TimedText, ... [recorded in
  <a href=
  "http://www.w3.org/2009/04/17-mediafrag-minutes.html#action06">http://www.w3.org/2009/04/17-mediafrag-minutes.html#action06</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> raphael to split
  the document and setup the diff feature of xmlspec [recorded in
  <a href=
  "http://www.w3.org/2009/04/17-mediafrag-minutes.html#action08">http://www.w3.org/2009/04/17-mediafrag-minutes.html#action08</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> raphaël to split
  the document and setup the diff feature of xmlspec [recorded in
  <a href=
  "http://www.w3.org/2009/04/17-mediafrag-minutes.html#action09">http://www.w3.org/2009/04/17-mediafrag-minutes.html#action09</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 1.135 (<a href=
    "http://dev.w3.org/cvsweb/2002/scribe/">CVS log</a>)<br />
    $Date: 2009/04/17 19:35:59 $
  </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: 1.135  of Date: 2009/03/02 03:52:20
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)

WARNING: Bad s/// command: s/<a href=
"http://respighi.elis.ugent.be/NinSunaWeb/NN">http://respighi.elis.ugent.be/NinSunaWeb/NN</a>
WARNING: Bad s/// command: s/http:\/\/respighi.elis.ugent.be\/NinSunaWeb/NN/g
Succeeded: s/aprse/parse/
Succeeded: s/ohw/how/
Succeeded: s/gree/agree/
Succeeded: s/: can be both/: both test harnesses can be/
Succeeded: s/then put into/... then put into/
Succeeded: s/req docs/REC docs/
Found Scribe: Raphael
Inferring ScribeNick: raphael
Found ScribeNick: raphael
Found ScribeNick: mhausenblas
Found ScribeNick: erik
Found ScribeNick: raphael
Found ScribeNick: davy
ScribeNicks: raphael, mhausenblas, erik, davy

WARNING: Replacing list of attendees.
Old list: Mediafrag Silvia Yves
New list: Mediafrag Yves

Default Present: Mediafrag, Yves
Present: Erik Davy Guillaume Michael Jack Conrad Raphael Frank_(observer) Yves_(remote) Silvia_(remote) Dave
Agenda: <a href=
"http://www.w3.org/2008/WebVideo/Fragments/wiki/ThirdF2FAgenda">http://www.w3.org/2008/WebVideo/Fragments/wiki/ThirdF2FAgenda</a>
Found Date: 17 Apr 2009
Guessing minutes URL: <a href=
"http://www.w3.org/2009/04/17-mediafrag-minutes.html">http://www.w3.org/2009/04/17-mediafrag-minutes.html</a>
People with action items: conrad erik jack l michael rapha raphael

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