Another abandoned server code base... this is kind of an ancestor of taskrambler.
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
 

1161 lines
45 KiB

<!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>