20-mediafrag-minutes.html 29.5 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 -- 20 Oct
  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>Media Fragments Working Group Teleconference</h1>

  <h2>20 Oct 2008</h2>

  <p><a href=
  'http://www.w3.org/2008/WebVideo/Fragments/wiki/FirstF2FAgenda'>Agenda</a></p>

  <p>See also: <a href=
  "http://www.w3.org/2008/10/20-mediafrag-irc">IRC log</a></p>

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

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

      <dd>Iles_C</dd>

      <dt>Regrets</dt>

      <dt>Chair</dt>

      <dd>Erik, Raphael</dd>

      <dt>Scribe</dt>

      <dd>Jack</dd>
    </dl>
  </div>

  <h2>Contents</h2>

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

      <ol>
        <li><a href="#item01">1. Round of introductions</a></li>

        <li><a href="#item02">2. Use Cases Discussion (Part
        1)</a></li>

        <li><a href="#item03">3. Use Case Discussion (Part
        2)</a></li>

        <li><a href="#item04">Media Delivery use case</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: 20 October
    2008</p>

    <p class='irc'>&lt;<cite>nessy</cite>&gt; Meeting openend
    9:08</p>

    <h3 id="item01">1. Round of introductions</h3>

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

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

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

    <p class='phone'><cite>Davy:</cite> also in Multimedia Lab,
    IBBT, Ghent (BE)</p>

    <p class='phone'><cite>Silvia:</cite> involved in MPEG-7,
    MPEG-21, developed Annodex (annotation format for ogg media
    files)<br />
    ... start my own start up for measuring the audience of video
    on the web + consultant for Mozilla<br />
    ... developped the TemporalURI specification, 6 years ago</p>

    <p class='phone'>Guillaume Olivrin, South Africa, focus on
    accessibility, how do you attach specific semantics to parts of
    media</p>

    <p class='phone'>Daniel Park, Samsung, co-chair of the Media
    Annotation, focus on IPTV (background in wireless
    networking)</p>

    <p class='phone'>Andy Heath, Open University, UK, background on
    e-learning, but develop far more general technologies, focus on
    accessibility</p>

    <p class='phone'><cite>scribe:</cite> experience in standards
    such as LOM, DC, SKORM</p>

    <p class='phone'>Colm Doyle: Blinkx</p>

    <p class='phone'>Larry Masinter: Adobe, experience in
    co-chairing HTTP group, focus on acquisition of metadata</p>

    <p class='phone'>Khang Cham, Samsung, focus on IPTV</p>

    <p class='phone'><cite>Yves:</cite> W3C team contact, expertise
    in protocols, web services</p>

    <p class='irc'>&lt;<cite>nessy</cite>&gt; <a href=
    "http://www.w3.org/2008/01/media-fragments-wg.html">http://www.w3.org/2008/01/media-fragments-wg.html</a></p>

    <p class='irc'>&lt;<cite>nessy</cite>&gt; ... working group
    charter</p>

    <p class='phone'><cite>Larry:</cite> important to define first
    requirements for why these URIs will be used for<br />
    ... it might happen that you can not satisfy all the
    requirements with a URI, don't put that out of scope now</p>

    <h3 id="item02">2. Use Cases Discussion (Part 1)</h3>

    <p class='phone'>Photo Use Case: <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/wiki/Use_Cases_%26_Requirements#Photobook_UC">
    http://www.w3.org/2008/WebVideo/Fragments/wiki/Use_Cases_%26_Requirements#Photobook_UC</a></p>

    <p class='phone'>Slides at: <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/meetings/2008-10-20-f2f_cannes/photobook_UC.pdf">
    http://www.w3.org/2008/WebVideo/Fragments/meetings/2008-10-20-f2f_cannes/photobook_UC.pdf</a></p>

    <p class='phone'>Erik goes through the slides</p>

    <p class='phone'><cite>Erik:</cite> take parts of images ...
    and assemble them together in a slideshow</p>

    <p class='phone'><cite>Guillaume:</cite> unclear the value of
    the fragments here<br />
    ... I understand fragment as taking a part of a large thing</p>

    <p class='phone'><cite>Larry:</cite> is it worth at all to look
    at Spatial URIs? Is it for doing partial retrieval?</p>

    <p class='phone'><cite>Raphael:</cite> mention maps
    applications</p>

    <p class='phone'><cite>Larry:</cite> but they are
    intereactive!</p>

    <p class='phone'><cite>Raphael:</cite> mention multi-resolution
    images, image industry has huge need and will to expose high
    resolution version of images</p>

    <p class='phone'><cite>Larry:</cite> they do have JPEG2000 and
    protocols</p>

    <p class='phone'><cite>Silvia:</cite> SMIL has ellaborate on
    the need for spatial fragments</p>

    <p class='phone'><cite>Jack:</cite> important needs in the SMIL
    community and SVG ... image maps, pan zoom, cropping</p>

    <p class='phone'><cite>Erik:</cite> continues the presentation,
    after temporally assemble parts of images into a slideshow,
    assemble two parts of an image into a new one (stich)<br />
    ... Existing technologies: RSS and Atom for the playlist
    generation<br />
    ... W3C SMIL: XML-based markup language, requires a SMIL
    player<br />
    ... MPEG-21: Part 17 for fragment identification of MPEG
    Ressources, client-side processing ... pseudo playlist<br />
    ... MPEG-A: MAF (Media Application Format) that combines MPEG
    technologies<br />
    ... XSPF (spiff): XML Shareable Playlist Format: Xiph
    Community<br />
    ... Discussion: is it out of scope or not? specific use cases
    around? other technologies around?</p>

    <p class='phone'><cite>Guillaume:</cite> unclear the value of
    the fragments here<br />
    ... I understand fragment as taking a part of a large thing</p>

    <p class='phone'><cite>Silvia:</cite> we are mainly looking at
    audio and videos files, but a video is a sequence of images</p>

    <p class='phone'><cite>Larry:</cite> there are different
    servers and clients</p>

    <p class='phone'><cite>Silvia:</cite> one way to look at a
    criteria is: is it a pure client-side issue or server-side +
    client-side problems?</p>

    <p class='phone'><cite>Larry:</cite> even if it is only a
    client-side issue, it might be worth to do some
    standardisation<br />
    ... the main point of still images fragment is the
    interactivity</p>

    <p class='phone'><cite>Raphael:</cite> is interactivity the key
    interest in spatial fragment</p>

    <p class='phone'><cite>Larry:</cite> there is a lot of work in
    this area, would recommend to focus on the temporal issue<br />
    ... it is also a good exercise to look at the out-of-scope use
    case, help to shape the scope</p>

    <p class='phone'><cite>Jack:</cite> URI is good because it is
    the web, the client is not necessarily aware of the time
    dimension<br />
    ... HTML has already a notion of Area, so don't encode it in a
    URI</p>

    <p class='phone'><cite>Larry:</cite> need to be carreful on
    URIs, resources, representations<br />
    ... example of an image: need to decode it, take the parts,
    re-encode it<br />
    ... JPEG2000 might have a direct way to do that</p>

    <p class='phone'><cite>Guillaume:</cite> create mosaic, collage
    of parts of media</p>

    <p class='phone'><cite>Yves:</cite> it depends if the
    transformation needs to be on the client or not</p>

    <p class='phone'><cite>Jack:</cite> be carreful, to not put SVG
    in a URI :-)<br />
    ... good balance on which processing can be on client side, and
    what is worth to put in a URL<br />
    ... is it better to have the processing in the URL?</p>

    <p class='phone'><cite>Erik:</cite> we question again the
    interest of the spatial fragment</p>

    <p class='phone'><cite>Silvia:</cite> is it a question of the
    size of the media? Large: worth to have fragment, Small: not
    worth</p>

    <p class='phone'><cite>Larry:</cite> define what do you mean by
    media<br />
    ... it is reasonable to limit yourself to videos</p>

    <p class='phone'><cite>Silvia:</cite> SMIL and Flash are
    interactive media, not necessarily one timeline<br />
    ... we focus on a resource with one timeline<br />
    ... there is a whole sweat of codecs issues</p>

    <p class='phone'><cite>Larry:</cite> define markers in
    videos</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; time... what is the
    reference of time for a video, embedded time code? 0 for the
    start?</p>

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

    <p class='phone'>Map Use Case: <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/wiki/Use_Cases_%26_Requirements/Map_Application_UC">
    http://www.w3.org/2008/WebVideo/Fragments/wiki/Use_Cases_%26_Requirements/Map_Application_UC</a></p>

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

    <p class='phone'><cite>Raphael:</cite> Map UC Description</p>

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

    <p class='phone'><cite>Raphael:</cite> Annotation is key</p>

    <p class='irc'>&lt;<cite>Kangchan</cite>&gt; Question : What is
    relation between Geolocation Working Group(with <a href=
    "http://www.w3.org/2008/geolocation/)">http://www.w3.org/2008/geolocation/)</a>
    and Web Map Services</p>

    <p class='phone'><cite>Raphael:</cite> UC examples using Yahoo,
    Google &amp; Microsoft</p>

    <p class='phone'><cite>Jack:</cite> what we see here are URI's
    for the applications, not images</p>

    <p class='phone'>Raphael will look deeper into different specs
    over the next couple of weeks for this Map UC</p>

    <p class='phone'>Davy &amp; Jack: is this a valid UC? will our
    spatial URL adressing scheme will be used by Maps
    Applications?</p>

    <p class='phone'><cite>Raphael:</cite> as Larry said this
    morning, out-of-scope UC's are valid to come up with our final
    WG's scope</p>

    <p class='irc'>&lt;<cite>guillaume</cite>&gt; Must document the
    out of scope UC&nbsp;to explain why it is out of scope.</p>

    <p class='phone'><cite>Sylvia:</cite> there might be a UC when
    we are talking about really large images (cfr. medical images
    in really high resolutions)<br />
    ... having a way to get a subpart of such a big image is nice
    to have, but implementation is something different ... a lot of
    complications, certainly on some server-side
    implimentations</p>

    <p class='phone'><cite>Guillaume:</cite> codec issues not to be
    underestimated, have a nice adressing scheme vs. server-side
    complexity</p>

    <p class='phone'><cite>Sylvia:</cite> should look further than
    just server-side complexity, solutions for certain codecs will
    come around eventually if needed</p>

    <p class='phone'><cite>Jack:</cite> pratical issues vs.
    fundamental issues have to be taken into account within this
    group<br />
    ... media fragments are needed because some things can not be
    expressed today</p>

    <p class='phone'><cite>Raphael:</cite> is it worth of having an
    overview of the TimedText WG?</p>

    <p class='irc'>&lt;<cite>nessy</cite>&gt; Guillaume: URI
    fragment identifier for text/plain: <a href=
    "http://www.ietf.org/rfc/rfc5147.txt">http://www.ietf.org/rfc/rfc5147.txt</a></p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; (multi-resolution
    formats, <a href=
    "http://en.wikipedia.org/wiki/FlashPix">http://en.wikipedia.org/wiki/FlashPix</a>
    is a good example of a single file containing multiple
    resolutions, maybe better than the map application)</p>

    <p class='phone'><cite>Raphael:</cite> Zoomify is good example
    of UC of very big images (life sciences) using fragments<br />
    ... task of this group to ensure interoperability of different
    standards? (eg. MPEG-21 URI to SVG)</p>

    <p class='phone'><cite>Sylvia:</cite> defining the mappings
    should be out-of-scope for this WG</p>

    <p class='phone'><cite>Jack:</cite> worthwile is testing our
    scheme to the others out there</p>

    <p class='phone'><cite>Sylvia:</cite> last thing to do &amp;
    should be straight forward by then if we did a good job</p>

    <p class='phone'><cite>Raphael:</cite> what about spatial
    dimension?</p>

    <p class='phone'><cite>Sylvia:</cite> temporal adressing need
    is biggest, but spatial adressing need is also valid</p>

    <h3 id="item03">3. Use Case Discussion (Part 2)</h3>

    <p class='phone'>Sylvia presenting the Media Annotation UC</p>

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

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Silvia: Annotation
    can be attached to the full media resource or to fragments of
    media resources</p>

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

    <p class='phone'><cite>Sylvia:</cite> annotations to fragment
    is relevant for this group</p>

    <p class='phone'><cite>Guillaume:</cite> can the structure of
    the video be represented in the URI</p>

    <p class='phone'><cite>Silvia:</cite> difference between the
    representation of the fragment and its semantics</p>

    <p class='irc'>&lt;<cite>spark3</cite>&gt; if necessary, what
    about adding a new UC (naming use case for fragment) into the
    Media Annotation WG UC ?</p>

    <p class='phone'><cite>Silvia:</cite> drawing on the board</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; Jack: there's only 1
    timeline for timed media</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; Jack: there's only 1
    coordinate system for spatial media</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; Jack: Annotation UC is
    important because we're reasoning on a higher abstraction
    level</p>

    <p class='phone'><cite>Jack:</cite> loves that use case since
    it is purely about fundamental description and indexing of a
    media</p>

    <p class='phone'><cite>Silvia:</cite> goes through the
    advantages of a possible URI scheme for media fragments<br />
    ... actually motivating the need for media fragments<br />
    ... shows the picture at <a href=
    "https://wiki.mozilla.org/Image:Video_Fragment_Linking.jpg">https://wiki.mozilla.org/Image:Video_Fragment_Linking.jpg</a><br />

    ... jumps into the track problems<br />
    ... there is actually 3 dimensions: space, time and track<br />
    ... temporalURI just deal with cropping, no track awareness</p>

    <p class='phone'><cite>Jack:</cite> rename this use case into
    'Anchoring'<br />
    ... annotation = RDF community<br />
    ... structuring = SMIL community</p>

    <p class='phone'><cite>Silvia:</cite> agree to rename it into
    Media Anchor Definition</p>

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

    <p class='phone'>Media Delivery Use Case: <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/wiki/Use_Cases_%26_Requirements#Media_Delivery_UC">
    http://www.w3.org/2008/WebVideo/Fragments/wiki/Use_Cases_%26_Requirements#Media_Delivery_UC</a></p>

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

    <p class='irc'>&lt;<cite>Yves</cite>&gt; Scribe: Jack</p>

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

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

    <h3 id="item04">Media Delivery use case</h3>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Davy going through
    the slide at: <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/meetings/2008-10-20-f2f_cannes/media_delivery_UC.pdf">
    http://www.w3.org/2008/WebVideo/Fragments/meetings/2008-10-20-f2f_cannes/media_delivery_UC.pdf</a></p>

    <p class='phone'><cite>Various:</cite> (discussing slide 3, #
    vs. ? or ,): Can we use # as the only user-visible marker and
    use http-ranges or something similar?</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Silvia drawing a
    communication channel between UA and servers</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Discussion about
    the use of the "hash" character</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Yves: use case is
    to extract a frame of a video, and creates a new image (so a
    new resource), use a '?'</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... use case is to
    keep the context, use a '#'</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Summary: there is
    use cases for both, should be further discussed tomorrow
    morning</p>

    <p class='phone'><cite>summary:</cite> there are use cases for
    both. We will get back to the subject tomorrow.</p>

    <p class='irc'>&lt;<cite>guillaume</cite>&gt; dejà vue</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Davy: explains the
    MPEG-21 Fragment identification</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... use of the '#',
    but no delivery protocol</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... mention also
    the proposal of Dave Singer: UA get first N bytes representing
    the headers with timing and bytes offset information of the
    media resource</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... goes through an
    explanation of MPEG-21: <a href=
    "http://www.w3.org/2008/WebVideo/Fragments/wiki/State_of_the_Art#MPEG-21_Part_17:_Fragment_Identification_of_MPEG_Resources_.28Davy_.2F_Silvia.29">
    http://www.w3.org/2008/WebVideo/Fragments/wiki/State_of_the_Art#MPEG-21_Part_17:_Fragment_Identification_of_MPEG_Resources_.28Davy_.2F_Silvia.29</a></p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... 4 schemes</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... ffp for the
    track</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... offset for
    bytes range</p>

    <p class='phone'><cite>all:</cite> discussing #mp() scheme</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... mp for
    specifying the temporal or spatial fragment (only for MPEG
    mime-type resources)</p>

    <p class='phone'><cite>sylvia:</cite> whoever controls the
    mimetype also controls what is after the # in a url</p>

    <p class='phone'><cite>Jack:</cite> is surprised, but
    pleasantly so.</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Davy: the 4th
    scheme is 'mask' (only for MPEG resources)</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Jack: seems they
    structure the video resource and point towards this
    structure</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Raphael: how many
    user agents can understand this syntax?</p>

    <p class='phone'><cite>all:</cite> none, that we know of</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Davy: i'm not aware
    of ... altough there is a referenced implementation</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Larry: http is not
    necessarily the best protocol to transport video</p>

    <p class='irc'>&lt;<cite>Yves</cite>&gt; in video, it depends
    if you want exact timing, control of the lag, and in that case
    HTTP is not the best choice</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Silvia: I would say
    that most of the videos is transported over http</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... RTP and RTSP
    have their own fragments, we should learn from them</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... if they do not
    satisfy all our requirements, we can feed them so they extend
    the use of fragments in these protocols</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Davy: goes through
    TemporalURI</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... this is the
    only that specifies a delivery protocol over http</p>

    <p class='phone'><cite>Silvia:</cite> Real used to allow
    something similar to temporal URLs</p>

    <p class='phone'><cite>Jack:</cite> thinks it may be part of
    the .ram files</p>

    <p class='phone'><cite>Guillaume:</cite> Flash allows doc
    author to export subparts by name, these can then be accessed
    with url#name</p>

    <p class='phone'><cite>Davy:</cite> continues with slide 6,
    http media delivery</p>

    <p class='irc'>&lt;<cite>guillaume</cite>&gt; Guillaume: Flash
    could also embed internal links in movie attached to certain
    frames. Once compiled with specific option, fragment of the
    Flash movie could be accessed using #</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Silvia: draw the
    four-way handshake</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... 1st exchange:
    User requests <a href=
    "http://www.example.com/resource.ogv#t=20-30">http://www.example.com/resource.ogv#t=20-30</a></p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... UA does a GET
    &lt;uri stripped of hash&gt;, Range: time 20-30</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... Server send
    back a Response 200, with the content-range: time 20-30 +
    content-type + ogg header + time-range bytes 50000-20000</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... (needs to
    create a new http header, 'time-range')</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Raphael: can we use
    content-range: bytes ... ?</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... UA does a GET
    &lt;URI strriped of the hash&gt;, Range x bytes: 5000-20000</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... Server send
    back a Response 200, with the content-range bytes + the cropped
    data</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Silvia: it is not
    implemented yet as far as I know</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... discussion
    based on a lot of discussions with proxies vendors</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Davy: could we
    apply the same four-way handshake with RTSP?</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... RTSP specifies
    a Range Header, similar to the HTTP byte range mechanism</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... RTSP could
    support temporal fragments by a two-way handshake (using Range
    header)</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... Problem:
    spatial fragments are not supported!</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Jack: the spatial
    problem is kind of orthogonal</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... the spatial
    fragment will not be about bytes range</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Davy: cropping is
    more complex in images</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Jack: you're right,
    I can create a non-continous quicktime movie</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; ... problem is it
    is not necessarily possible to generate a byte range from a
    time range</p>

    <p class='irc'>&lt;<cite>raphael</cite>&gt; Silvia: a single
    byte range</p>

    <p class='phone'><cite>all:</cite> the non-contiguous ranges
    may occur more often than we like. But maybe<br />
    ... we can get away with ignoring them (because all relevant
    formats also have a contiguous form).<br />
    ... need to discuss after the break.</p>

    <p class='phone'><cite>raphael:</cite> suggest coffee break</p>

    <p class='irc'>&lt;<cite>guillaume</cite>&gt; or need to
    coalesce</p>

    <p class='phone'><cite>Larry:</cite> please decouple
    representation of how you refer to fragments form he
    implementations<br />
    ... Also think about embedded metadata: if the original has a
    copyright statement, do you get it wth every fragment?</p>

    <p class='phone'><cite>Sylvia:</cite> (on prev subject):
    wonders whether http can do multiple byte ranges</p>

    <p class='phone'><cite>Larry:</cite> yes, I think so, with
    multipart</p>

    <p class='irc'>&lt;<cite>erik</cite>&gt; rssagent, draft
    minutes</p>

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

    <p class='phone'>Media Linking UC</p>

    <p class='phone'>raphael discusses the description written by
    Michael on the wiki</p>

    <p class='phone'><cite>scribe:</cite> 3 things: bookmarking,
    playlists, and interlinking multimedia</p>

    <p class='phone'><cite>silvia:</cite> definition of playlists
    is out of scope</p>

    <p class='phone'><cite>guillaume:</cite> playlist is about
    presentation</p>

    <p class='phone'><cite>raphael:</cite> regarding interlinked:
    temporal URIs can be described in RDF (RDF doc describing an
    audio file)<br />
    ... difference between URI and RDF (or SMIL, or ...): you need
    to parse the metadata<br />
    ... RDF description of time segment could be replaced by a
    temporal URI</p>

    <p class='phone'><cite>silvia:</cite> interlinking multimedia
    is already covered in other UCs</p>

    <p class='phone'>Video Browser UC</p>

    <p class='phone'><cite>silvia:</cite> large media files
    introduces special challenges<br />
    ... requirement for server-side processing<br />
    ... dynamic creation of thumbnails through URI mechanism</p>

    <p class='phone'><cite>guillaume:</cite> link to PNG or
    GIF<br />
    ... provide a preview function of the resource<br />
    ... trivial: get all the I-frames of a video resource<br />
    ... use them as thumbs<br />
    ... thumbnail extraction is quite easy</p>

    <p class='phone'>silvia, jack: not so trivial, might be
    processing-intensive</p>

    <p class='phone'><cite>silvia:</cite> it should be possible to
    point to one single frame with the URI scheme</p>

    <p class='phone'><cite>jack:</cite> URI scheme should not know
    that frame is 'the' thumbnail</p>

    <p class='phone'><cite>guillaume:</cite> you can have multiple
    thumbs per resource</p>

    <p class='phone'><cite>raphael:</cite> URI scheme can point to
    a frame, but does not have knowledge about thumbs<br />
    ... should we be able to address in terms of frames?</p>

    <p class='phone'><cite>guillaume:</cite> no, too
    coding-specific</p>

    <p class='phone'><cite>silvia:</cite> previews of images?<br />
    ... preview is then a lower resolution image</p>

    <p class='phone'><cite>guillaume:</cite> that is
    processing<br />
    ... mostly, previews are already part of the media
    resource<br />
    ... hence lower image resolutions are out of scope</p>

    <p class='phone'><cite>jack:</cite> not too far?<br />
    ... is a preview embedded in a resource still a fragment?</p>

    <p class='phone'><cite>guillaume:</cite> compare it with
    tracks<br />
    ... preview is just another track</p>

    <p class='phone'><cite>raphael:</cite> we put this in mind and
    make a decision later</p>

    <p class='phone'><cite>silvia:</cite> previews are another sort
    of tracks</p>

    <p class='phone'><cite>raphael:</cite> should we also to be
    able to address metadata within the headers?</p>

    <p class='phone'><cite>silvia:</cite> it is not a common
    property of all the formats to have previews, therefore, it is
    not a candidate to be standardized</p>

    <p class='phone'><cite>raphael:</cite> after first fase of the
    WG: report the current limitations<br />
    ... and wait for feedback</p>

    <p class='phone'>Moving Point Of Interest UC</p>

    <p class='phone'><cite>raphael:</cite> complex UC<br />
    ... should be for the second phase</p>

    <p class='phone'><cite>jack:</cite> if this ever to be going to
    used at server-side?<br />
    ... if not, it is out of scope</p>

    <p class='phone'><cite>raphael:</cite> you can share the link
    of the moving region</p>

    <p class='phone'><cite>erik:</cite> delivery to mobile devices
    is a use case introduced by the public flemish broadcaster</p>

    <p class='phone'><cite>jack:</cite> there is no reason to use
    URIs for that purpose, use metadata</p>

    <p class='phone'><cite>raphael:</cite> it is like concatenating
    spatial fragments over time</p>

    <p class='phone'><cite>guillaume:</cite> we are addressing
    points over space or time</p>

    <p class='phone'><cite>raphael:</cite> refer to HTML image
    maps<br />
    ... region, interval can be defined by a combination of
    points<br />
    ... you need more than one point</p>

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

    <p class='phone'><cite>raphael:</cite> we will discuss this
    tomorrow</p>
  </div>

  <h2><a name="ActionSummary" id="ActionSummary">Summary of Action
  Items</a></h2><!-- Action Items -->
  [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: 2008/10/26 10:50:37 $
  </address>

</body>
</html>