16-mediaann-minutes.html 20.3 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>mawg -- 16 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="mawg" 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>mawg</h1>

  <h2>16 Apr 2009</h2>

  <p><a href=
  'http://www.w3.org/2008/WebVideo/Annotations/wiki/Meeting_Agenda_(In_Progress)'>
  Agenda</a></p>

  <p>See also: <a href=
  "http://www.w3.org/2009/04/16-mediaann-irc">IRC log</a></p>

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

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

      <dd>Doug_Schepers, Felix</dd>

      <dt>Regrets</dt>

      <dt>Chair</dt>

      <dd>daniel</dd>

      <dt>Scribe</dt>

      <dd>tbd, wbailer</dd>
    </dl>
  </div>

  <h2>Contents</h2>

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

      <ol>
        <li><a href="#item01">mxm introduction</a></li>

        <li><a href="#item02">review of mapping table</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>fsasaki</cite>&gt; scribe: tbd</p>

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

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

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Date: 16 April
    2009</p>

    <p class='irc'>&lt;<cite>vrodrgue2</cite>&gt; I am vroddon2 in
    skype</p>

    <p class='irc'>&lt;<cite>daniel</cite>&gt; felix, can you hear
    our voices ?</p>

    <p class='irc'>&lt;<cite>fsasaki</cite>&gt; very hard to
    hear</p>

    <p class='irc'>&lt;<cite>shepazu</cite>&gt; XMP also has broad
    support on deployed devices</p>

    <p class='irc'>&lt;<cite>shepazu</cite>&gt; like cameras,
    videocams, etc.</p>

    <p class='irc'>&lt;<cite>daniel</cite>&gt; canonical processes
    discussion proposed by Venonique</p>

    <p class='irc'>&lt;<cite>daniel</cite>&gt; dave is talking
    about some issues and questions on UC&amp;Req document.</p>

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

    <p class='phone'><cite>dave:</cite> cross site scripting is an
    security issue for metadata access<br />
    ... mention that this is an issue, work on solution with other
    WGs<br />
    ... issue: are annotations timed or not?<br />
    ... applys to different type of metadata: content description,
    rights, etc.<br />
    ... metadata global to media item might be sufficient to start,
    but time-dependent metadata needs to be considered</p>

    <p class='irc'>&lt;<cite>daniel</cite>&gt; yeah, felix, the
    whether is very nice</p>

    <p class='irc'>&lt;<cite>daniel</cite>&gt; ..:-)</p>

    <p class='phone'><cite>joakim:</cite> what is the media object?
    is a fragment a media object<br />
    ... needs to be defined</p>

    <p class='phone'><cite>felix:</cite> this is an issue, do we
    need to solve it? could timed text approach be used for
    that</p>

    <p class='phone'><cite>dave:</cite> time reference should be
    supported in api</p>

    <p class='phone'><cite>joakim:</cite> doesn't media fragments
    solve that? uri describes fragment<br />
    ... bring up question in joined section</p>

    <p class='irc'>&lt;<cite>fsasaki</cite>&gt; felix: media
    fragments group mechanism might be helpful here</p>

    <p class='phone'><cite>dave:</cite> use case "Access via web
    client to metadata in heterogeneous formats"<br />
    ... to which collection do the queries refer to?</p>

    <p class='irc'>&lt;<cite>vmalais</cite>&gt; <a href=
    "http://dev.w3.org/cvsweb/~checkout~/2008/video/mediaann/mediaont-req/mediaont-req.html?rev=1.57&amp;content-type=text/html;%20charset=utf-8">
    http://dev.w3.org/cvsweb/~checkout~/2008/video/mediaann/mediaont-req/mediaont-req.html?rev=1.57&amp;content-type=text/html;%20charset=utf-8</a></p>

    <p class='phone'><cite>dave:</cite> specify what goes along the
    arrows in the diagram in section3</p><a name="action01" id=
    "action01"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> felix to clarify arrow + descriptions
    for the diagram in section 3 of UC &amp; req dcouments
    [recorded in <a href=
    "http://www.w3.org/2009/04/16-mediaann-minutes.html#action01">http://www.w3.org/2009/04/16-mediaann-minutes.html#action01</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Created ACTION-101
    - Clarify arrow + descriptions for the diagram in section 3 of
    UC &amp; req dcouments [on Felix Sasaki - due 2009-04-23].</p>

    <p class='phone'><cite>florian:</cite> uc 5.4 refers accessing
    the api from a search engine</p>

    <p class='phone'><cite>joakim:</cite> check if attributes
    mentioned in attributes are covered by properties in mapping
    table</p><a name="action02" id="action02"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> veroniqueM to update canonical
    processes use cases [recorded in <a href=
    "http://www.w3.org/2009/04/16-mediaann-minutes.html#action02">http://www.w3.org/2009/04/16-mediaann-minutes.html#action02</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Sorry, couldn't
    find user - veroniqueM</p><a name="action03" id="action03"></a>

    <p class='irc'>&lt;<cite>scribe</cite>&gt;
    <strong>ACTION:</strong> vmalais to update canonical processes
    use cases [recorded in <a href=
    "http://www.w3.org/2009/04/16-mediaann-minutes.html#action03">http://www.w3.org/2009/04/16-mediaann-minutes.html#action03</a>]</p>

    <p class='irc'>&lt;<cite>trackbot</cite>&gt; Created ACTION-102
    - Update canonical processes use cases [on Véronique Malaisé -
    due 2009-04-23].</p>

    <p class='irc'>&lt;<cite>daniel</cite>&gt; coffee break</p>

    <p class='irc'>&lt;<cite>daniel</cite>&gt; and get back to the
    MXM discussion...</p>

    <h3 id="item01">mxm introduction</h3>

    <p class='phone'>wonsuk presents short overview of mpeg
    extensible middleware</p>

    <p class='phone'><a href=
    "http://mxm.wg11.sc29.org/">http://mxm.wg11.sc29.org/</a></p>

    <p class='irc'>&lt;<cite>daniel</cite>&gt; resume the meeting
    with MXM introduction by Wonsuk</p>

    <p class='phone'>wonsuk believes there is synergy between
    implementation of generic metadata engine in mxm and mawg</p>

    <p class='phone'><cite>joakim:</cite> define basic set of
    entities and their mappings to different formats<br />
    ... in future, further formats could be added<br />
    ... define basic set of entities</p>

    <p class='phone'><cite>veronique:</cite> from which formats to
    take definition of these entities</p>

    <p class='phone'><cite>joakim:</cite> entities are internal to
    ontology<br />
    ... few, generic entities should be sufficient</p>

    <p class='phone'><cite>jean-pierre:</cite> people will map to
    other formats - what to map to?</p>

    <p class='phone'><cite>florian:</cite> a defined reference is
    necessary<br />
    ... group and classify properties</p>

    <p class='phone'><cite>jean-pierre:</cite> how are entities
    from different formats related<br />
    ... define metamodel?</p>

    <p class='phone'><cite>ruben:</cite> define minimum set of core
    terms</p>

    <p class='phone'><cite>jean-pierre:</cite> went through this
    discussion =&gt; ebu core<br />
    ... europeana: reduce to 15 elements, then enriched</p>

    <p class='phone'><cite>veronique:</cite> start with core set,
    xmp also has core set going back to DC + enrichments</p>

    <p class='phone'><cite>ruben:</cite> ID3 is good example for
    basic metadata set</p>

    <p class='phone'><cite>joakim:</cite> publish set of general
    entities, make people think in terms of these entities</p>

    <p class='phone'><cite>dave:</cite> we'd all like a metadata
    system with clear semantics, widely used<br />
    ... don't know how to get it</p>

    <p class='phone'><cite>veronique:</cite> enforcing a format is
    not possible, mapping is the only option</p>

    <p class='phone'><cite>florian:</cite> mapping always means
    loss of information</p>

    <p class='phone'><cite>victor:</cite> mapping using semantic
    technologies can provide richer mapping than other
    apporaches<br />
    ... if set of entities is published in w3c recommendation
    people might be encouraged to use it</p>

    <p class='phone'><cite>florian:</cite> problem is different
    semantics in enrichment of standards<br />
    ... build simple core ontology, can be extended to build
    complex annotation</p>

    <p class='phone'><cite>jean-pierre:</cite> kind of combination
    of dc and mpeg-7 reflects interest of people in wg, but not of
    interest</p>

    <p class='irc'>&lt;<cite>fsasaki</cite>&gt; +1 to a simple set
    of properties</p>

    <p class='phone'><cite>jean-pierre:</cite> unthinkable
    combination<br />
    ... we have even not agreed on which format to use for
    describing mappings<br />
    ... most search engines make search based on brute force</p>

    <p class='phone'><cite>veronique:</cite> there are swoogle type
    of search engines</p>

    <p class='phone'><cite>jean-pierre:</cite> even if there's an
    api for mapping different formats- will it be used</p>

    <p class='phone'><cite>joakim:</cite> assemble domain
    knowledge, not filter it</p>

    <p class='phone'><cite>jean-pierre:</cite> work would be simple
    if all formats would be described already in rdf</p>

    <p class='phone'><cite>joakim:</cite> thought that 1:1 mapping
    of properties makes work easier</p>

    <p class='phone'><cite>veronique:</cite> adding entities and
    their relations would mean defining another ontology</p>

    <p class='phone'><cite>joakim:</cite> grouping properties like
    in xmp</p>

    <p class='phone'><cite>florian:</cite> define ontology, provide
    browsing properties without the need to understand owl</p>

    <p class='phone'><cite>veronique:</cite> develop ontology for
    grouping, avoid define another metadata format</p>

    <p class='phone'><cite>dave:</cite> no reason to prefer certain
    format, define properties and mappings to existing
    formats<br />
    ... not really defining a new ontology<br />
    ... not saying that tags we are mapping to are equivalent</p>

    <p class='phone'><cite>veronique:</cite> do people agree to
    define basic set of terms?</p>

    <p class='phone'><cite>jean-pierre:</cite> set of basic terms
    or superset?</p>

    <p class='phone'><cite>dave:</cite> subset of what you
    reasonable want and expect to get</p>

    <p class='irc'>&lt;<cite>fsasaki</cite>&gt; +1 to dave</p>

    <p class='phone'><cite>joakim:</cite> isn't that just dublinc
    core?</p>

    <p class='phone'><cite>veronqiue:</cite> is this really all we
    want?</p>

    <p class='phone'><cite>florian:</cite> in an image, one wants
    to annotate content</p>

    <p class='phone'><cite>ruben:</cite> doing better and more
    precise might be understand as just another ontology</p>

    <p class='phone'><cite>felix:</cite> what we're doing is
    developing dc for media<br />
    ... not bad to refer to dc<br />
    ... mappings to other formats not availble yet<br />
    ... lots of mappings available for dc elements in mapping
    table, getting thinner for other elements<br />
    ... concern that group is doing too much and progressing too
    slow<br />
    ... doing something simple for v1 and more complex for v2 might
    be a useful approach</p>

    <p class='phone'><cite>daniel:</cite> make first basic version,
    gather feedback and improve based on feedback</p>

    <p class='phone'>daniel reminds of wg charter</p>

    <p class='phone'><cite>jean-pierre:</cite> in a simple
    approach, we define a list of terms (might be some as dc or
    xmp)<br />
    ... let's see if we agree on list of terms<br />
    ... define list of entities and mapping, including mapping dc
    and xmp to that list</p>

    <p class='phone'><cite>joakim:</cite> everyone agrees?</p>

    <p class='phone'>no disagreement</p>

    <p class='phone'><cite>jean-pierre:</cite> subclasses could be
    useful</p>

    <p class='phone'><cite>florian:</cite> would increase semantic
    expressiveness</p>

    <h3 id="item02">review of mapping table</h3>

    <p class='phone'><cite>joakim:</cite> level of conformance of
    implementations to be defined</p>

    <p class='phone'><cite>jean-pierre:</cite> implementation
    cannot mean change implementations</p>

    <p class='phone'><cite>joakim:</cite> have several levels of
    conformance?</p>

    <p class='irc'>&lt;<cite>daniel</cite>&gt; implementation
    volunteering (in this room): Joakim, Wonsuk, Werner</p>

    <p class='phone'><cite>jean-pierre:</cite> api for mapping =
    conformance</p>

    <p class='phone'><cite>joakim:</cite> what about proprietary
    formats of service providers</p>

    <p class='phone'><cite>jean-pierre:</cite> format should not be
    used as data model</p>

    <p class='phone'><cite>joakim:</cite> there are specific
    formats in business domains<br />
    ... use this in generic deployment environment</p>

    <p class='phone'><cite>jean-pierre:</cite> boundaries between
    delivery channels are disappearing<br />
    ... don't want to reinvent how to describe broadcast
    content</p>

    <p class='phone'><cite>dave:</cite> if there is a
    recommendation of comment set, why would broadcasters not use
    it?</p>

    <p class='phone'><cite>jean-pierre:</cite> broadcasters would
    keep their formats and use mappings</p>

    <p class='phone'><cite>dave:</cite> if there's a recommendation
    for set of properties, they would try to conform (in their
    formats)</p>

    <p class='phone'><cite>florian:</cite> common set of entities
    is needed for querying across several formats</p>

    <p class='phone'><cite>jean-pierre:</cite> content provider
    would not implement for mapping</p>

    <p class='phone'><cite>veronique:</cite> at least not large
    providers</p>

    <p class='phone'><cite>jean-pierre:</cite> we should not reject
    any particular type of metadata<br />
    ... for the time being we have a restricted set of formats for
    which we have defined mappings<br />
    ... everyone should be able to map their format<br />
    ... restrictions may come from the way the api is written</p>

    <p class='phone'><cite>veronique:</cite> skos might not be
    sufficient for describing the mappings</p>

    <p class='phone'><cite>ruben:</cite> if the internal format is
    hidden to the user, why standardise the mappings?<br />
    ... api only is visible</p>

    <p class='phone'><cite>jean-pierre:</cite> like dc, this would
    mean publishing list of terms</p>

    <p class='phone'><cite>florian:</cite> why should mapping rules
    be standardised?</p>

    <p class='phone'><cite>ruben:</cite> mapping can be useful, but
    can be informative</p>

    <p class='phone'><cite>jean-pierre:</cite> mapping is mechanism
    to discover the set of entities<br />
    ... but mappings need not be normative<br />
    ... simple list of properties is not incompatible with more
    complex ontology</p>

    <p class='phone'><cite>felix:</cite> simple approach is
    good<br />
    ... description of relation to other formats is very
    valuable<br />
    ... continue review of mappings</p>

    <p class='phone'><cite>veronique:</cite> would api be
    implementation of mappings?</p>

    <p class='phone'><cite>jean-pierre:</cite> is there need for an
    api?<br />
    ... if format in which to publish information is defined</p>

    <p class='phone'><cite>felix:</cite> api is useful for making
    mapping testable</p>

    <p class='phone'><cite>joakim:</cite> does that require solving
    data type issue of return values?</p>

    <p class='phone'>felix, dave: not necessary for making testable
    assertions</p>

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

    <p class='phone'><cite>veronique:</cite> list of terms, when to
    define?</p>

    <p class='phone'><cite>joakim:</cite> now<br />
    ... column with names of entities will be added to mapping
    table<br />
    ... then review and agree on format reviews</p>

    <p class='phone'><cite>veronique:</cite> are mappings one way
    or bidirectional?</p>

    <p class='phone'><cite>dave:</cite> one way only</p>

    <p class='phone'><cite>victor:</cite> refine mappings, specify
    transitivity, conditions under which properties are equivalent,
    etc<br />
    ... precising the relationship further<br />
    ... cardinalities<br />
    ... priorities in case of several mappings</p>

    <p class='phone'><cite>joakim:</cite> refining further would be
    necessary for supporting setting scenario</p>

    <p class='phone'><cite>dave:</cite> setting raises problem of
    value ranges of different formats</p>

    <p class='phone'><cite>joakim:</cite> future version of mapping
    table with refined semantics</p>

    <p class='phone'><cite>dave:</cite> choose terms that can be
    read from many formats</p>

    <p class='phone'><cite>jean-pierre:</cite> there might be terms
    that are useful but not (yet) widely supported</p>
  </div>

  <h2><a name="ActionSummary" id="ActionSummary">Summary of Action
  Items</a></h2><!-- Action Items -->
  <strong>[NEW]</strong> <strong>ACTION:</strong> felix to clarify
  arrow + descriptions for the diagram in section 3 of UC &amp; req
  dcouments [recorded in <a href=
  "http://www.w3.org/2009/04/16-mediaann-minutes.html#action01">http://www.w3.org/2009/04/16-mediaann-minutes.html#action01</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> veroniqueM to
  update canonical processes use cases [recorded in <a href=
  "http://www.w3.org/2009/04/16-mediaann-minutes.html#action02">http://www.w3.org/2009/04/16-mediaann-minutes.html#action02</a>]<br />

  <strong>[NEW]</strong> <strong>ACTION:</strong> vmalais to update
  canonical processes use cases [recorded in <a href=
  "http://www.w3.org/2009/04/16-mediaann-minutes.html#action03">http://www.w3.org/2009/04/16-mediaann-minutes.html#action03</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: 2009/04/16 10:46:04 $
  </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)

Succeeded: s/provideers/providers/
Succeeded: s/writter/written/
Found Scribe: tbd
Found Scribe: wbailer
Inferring ScribeNick: wbailer
Scribes: tbd, wbailer
Default Present: Doug_Schepers, Felix
Present: Doug_Schepers Felix

WARNING: Fewer than 3 people found for Present list!

Agenda: <a href=
"http://www.w3.org/2008/WebVideo/Annotations/wiki/Meeting_Agenda_(In_Progress)">http://www.w3.org/2008/WebVideo/Annotations/wiki/Meeting_Agenda_(In_Progress)</a>
Found Date: 16 Apr 2009
Guessing minutes URL: <a href=
"http://www.w3.org/2009/04/16-mediaann-minutes.html">http://www.w3.org/2009/04/16-mediaann-minutes.html</a>
People with action items: felix veroniquem vmalais

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