W3C home > Mailing lists > Public > public-media-annotation@w3.org > January 2010

minutes of 2010-01-26 teleconference

From: RaphaŽl Troncy <raphael.troncy@cwi.nl>
Date: Tue, 26 Jan 2010 14:07:23 +0100
Message-ID: <4B5EE90B.5090904@cwi.nl>
To: Media Annotation <public-media-annotation@w3.org>
Dear all,

These are the minutes in plain text of today's telecon. No formatting 



* Now talking in #mediaann
* Joins: florian (florian@
* Joins: pchampin (52e02448@
* Joins: tobias (5b0842ca@
* Joins: daniel (d25e2959@
* Joins: tmichel (tmichel@mcclure.w3.org)
* Joins: Zakim (rrs-bridgg@
<florian> Zakim, this is MAWG
<Zakim> ok, florian; that matches IA_MAWG()7:00AM
* Joins: joakim (jsderber@
<Zakim> + +
<Zakim> + +95177aabb
<pchampin> zakim, aabb is me
<Zakim> +pchampin; got it
<pchampin> *zakim, mute me
<pchampin> zakim, mute me
<Zakim> pchampin should now be muted
<Zakim> +joakim
<Zakim> +florian
<florian> Zakim, pick a victim
<Zakim> Not knowing who is chairing or who scribed recently, I propose 
* Joins: fsasaki (chatzilla@
* fsasaki zakim, code?
* Zakim saw 6294 (tel:+1.617.761.6200 tel:+ 
tel:+44.117.370.6152) given for the conference code, fsasaki
* Joins: veroniqueM (82251d84@
* florian veronique: will you join the telecon?
<Zakim> +??P5
* fsasaki zakim, ??P5 is Felix
* Zakim +Felix; got it
* fsasaki zakim, mute Felix
* Zakim Felix should now be muted
<Zakim> +[VrijeUni]
<Zakim> + +40.16.aacc
* daniel Veronique is a scribe today. are you ready ?
* veroniqueM yes
<veroniqueM> zakim, [VrijeUni] is me
<Zakim> +veroniqueM; got it
<veroniqueM> ITEM: approval of previous minutes
<pchampin> scribe: veroniqueM
<veroniqueM> ITEM: [a] Agenda review and approval of previous minutes 
<veroniqueM> item: selection of scribe for next telecon
<pchampin> I volunteer
<veroniqueM> Daniel: any volunteer?
* florian let?s pick a victim ;)
<pchampin> zakim, unmute me
<Zakim> pchampin should no longer be muted
<veroniqueM> pchampin volunteered
<pchampin> zakim, mute me
<Zakim> pchampin should now be muted
<veroniqueM> Daniel: will update the scribe list for next meeting
<veroniqueM> item: invited experts
<veroniqueM> Daniel: are you on the phone?
<veroniqueM> ... I will send an email to them, on the mailing list
<veroniqueM> ... to present the group
<veroniqueM> ... and will invite them to present themselves
<veroniqueM> item: action items
* fsasaki zakim, unmute Felix
* Zakim Felix should no longer be muted
<veroniqueM> http://www.w3.org/2008/WebVideo/Annotations/track/actions/open
<veroniqueM> http://www.w3.org/2008/WebVideo/Annotations/track/actions/124
* fsasaki zakim, mute Felix
* Zakim Felix should now be muted
<veroniqueM> Felix: no update
* fsasaki zakim, who is here?
* Zakim sees on the phone: ??P0, +, pchampin (muted), 
joakim, florian, Felix (muted), veroniqueM, +40.16.aacc
* Zakim sees on irc: veroniqueM, fsasaki, joakim, Zakim, tmichel, 
daniel, tobias, pchampin, florian, raphael, trackbot
* veroniqueM what was the command for keeping an action open?
* fsasaki you have to do nothing
* veroniqueM ok, nice :)
* fsasaki zakim, unmute Felix
* Zakim Felix should no longer be muted
<fsasaki> felix: 175 is part of 124
* fsasaki zakim, mute Felix
* Zakim Felix should now be muted
<veroniqueM> action items 135-171 pending, Veronique will check out the 
new Ontology document once it is published and do the updates required 
in the action items
* trackbot noticed an ACTION. Trying to create it.
<trackbot> Sorry, couldn't find user - items
<veroniqueM> close action 175
* fsasaki veronique, I jsut closed 175
<veroniqueM> http://www.w3.org/2008/WebVideo/Annotations/track/actions/189
<Zakim> + +49.300.aadd
* veroniqueM thanks Felix :) a bit overwhelmed by the speed here :)
<raphael> zakim, aadd is me
<Zakim> +raphael; got it
http://www.w3.org/2008/WebVideo/Annotations/track/actions/189 still pending
* raphael zakim, mute me
* Zakim raphael should now be muted
<tobias> Zakim, ??P0 is me
<Zakim> +tobias; got it
<florian> zakim, who is here?
<Zakim> On the phone I see tobias, +, pchampin (muted), 
joakim, florian, Felix (muted), veroniqueM, +40.16.aacc, raphael (muted)
<Zakim> On IRC I see veroniqueM, fsasaki, joakim, Zakim, tmichel, 
daniel, tobias, pchampin, florian, raphael, trackbot
* fsasaki cool! :)
* pchampin veronique: scribing tip: if you just type "action <number>", 
some bot will write the label of the action
* pchampin ... or may be "action-<number>"...
* veroniqueM what is the closing action item command?
* veroniqueM I did not have the time to get my papers here for the teleconf
<daniel> 204/205/206/207 are closed
<joakim> close ACTION-40
* trackbot attempting to close ACTION-40.
<trackbot> ACTION-40 to send around an exemple of what she means with 
"plugging into complex model" closed
<raphael> Trackbot hasn't been launched ?
* florian action-xy: closed
<raphael> there is no rrsagent !
<raphael> trackbot, start telecon
* trackbot is starting a teleconference
<trackbot> RRSAgent, make logs public
<trackbot> Zakim, this will be MAWG
<Zakim> ok, trackbot, I see IA_MAWG()7:00AM already started
<trackbot> Meeting: Media Annotations Working Group Teleconference
<trackbot> Date: 26 January 2010
* veroniqueM ok,was trying the other way around :)
<veroniqueM> action-204: closed
* trackbot attempting to add comment notes to ACTION-204.
<trackbot> ACTION-204 give informations about the workshop in korea 
notes added
<trackbot> If you meant to close ACTION-204, please use 'close ACTION-204'
<veroniqueM> close ACTION-204
* trackbot attempting to close ACTION-204.
<trackbot> ACTION-204 give informations about the workshop in korea closed
<joakim> chair:  Daniel
* florian sorry vero :( thought it was the other way
<veroniqueM> close ACTION-205
* trackbot attempting to close ACTION-205.
<trackbot> ACTION-205 resend the email with the updated ontology 
document to the editorial team closed
<joakim> scribe: Veronique
<veroniqueM> close ACTION-206
* trackbot attempting to close ACTION-206.
<trackbot> ACTION-206 new F2F near the MPEG Meeting (mpeg in geneva: 
10/07/26-30 & mpeg in dresden: 10/04/19-23) closed
<veroniqueM> close ACTION-207
* trackbot attempting to close ACTION-207.
<trackbot> ACTION-207 Fish for good ideas in the minutes and collect 
them in the wiki about the test suite closed
<veroniqueM> item: mapping table
<daniel> JP, Joakim, Chris, Tobias are done for the mapping
<veroniqueM> Joakim, Tobias, Felix, JP, Chris sent their mappings
<veroniqueM> Joakim: Victor tbd
<veroniqueM> Daniel: Wonsuk tbd
<raphael> rrsagent, draft minutes
<joakim> use this link!
<raphael> Thierry, what is going on with RRSAgent ? Why cannot we invite 
him to this channel ? How the logs will be generated ?
<veroniqueM> action: daniel to check out the status of the mapping table
* trackbot noticed an ACTION. Trying to create it.
<trackbot> Sorry, couldn't find user - daniel
<veroniqueM> Joakim: please remember that the update concerns both the 
order of the properties and the syntactic type of the mapping
<joakim> The order of the attributes should be the same as: 
<raphael> Could someone told me what is the output of the action 206 ? 
for closing it, what is the decision made regarding dates and location 
of next f2f ?
* veroniqueM Daniel sent an email to the mailing list about this
<raphael> URI pointer ?
* veroniqueM just checking
<veroniqueM> item: ontology document
<raphael> The last email from Daniel this morning is about voting, right ?
<raphael> so no decision yet regarding dates and location ?
<joakim> We will come back to 206
<veroniqueM> the new version will be online soon, and Veronique will 
check out what is ledt to be changed
<veroniqueM> no notice from editor team so far
-> [Thierry] PING
<veroniqueM> item: api document
* raphael Thierry: could you please look at what is happening with 
rrsagent ?
-> [tmichel] PING
<veroniqueM> Joakim: wants to discuss the structure of return object, 
but hard to discuss while Chris is not on the call
<tmichel> Yes I am looking at the issue and I don not understand why 
RRSagent is not responding
* Joins: wonsuk (wslee@
<veroniqueM> item: testsuite
<daniel> http://www.w3.org/2008/WebVideo/Annotations/wiki/TestSuite
* wonsuk hi all sorry for late because of business trip.
<veroniqueM> Daniel: Florian made a summary of previous discussions
* Joins: wbailer (8fe06491@
* veroniqueM @Raph F2F is in Korea feb 23-25
* veroniqueM more info on wiki but can't check out the uri while 
scribing, sorry
<tmichel> I acn't have RRS agent to work using /invite RRSAgent #mediaann
* raphael ok, I'm asking for the next next one ... Dresden, Vienna, etc. 
... the action 206 now closed !
<veroniqueM> Joakim: this should be the main topic of this meeting, the 
definition of a testsuite
<tmichel> invite RRSAgent #mediaann
<raphael> tmichel, I did the same, but irc told me no such nick/user
<veroniqueM> Florian: we have a few good examples for test cases
<tmichel> Zakim, who is here?
<Zakim> On the phone I see tobias, +, pchampin (muted), 
joakim, florian, Felix (muted), veroniqueM, +40.16.aacc, raphael (muted)
<Zakim> On IRC I see wbailer, wonsuk, veroniqueM, fsasaki, joakim, 
Zakim, tmichel, daniel, tobias, pchampin, florian, raphael, trackbot
<veroniqueM> ... we want to have a behavior for structured/unstructured 
<tmichel> rrsagent, help
* raphael lol thierry
<veroniqueM> ...these examples could be a starting point for this, with 
the examples/questions from the API document
<pchampin> zakim, mute me
<Zakim> pchampin was already muted, pchampin
<tmichel> rrsagent, draft minutes
<veroniqueM> ... connection lost error, api-based error
* raphael thinking RRSAgent is ill today
<veroniqueM> ...someone could be in charge of one example and 
implement/test it, then we could have a good coverage
* florian :)
<veroniqueM> Joakim: two testsuites? One for ontology document, one for 
API document
<veroniqueM> Floriam/Daniel: yes
<pchampin> what would a test suite for the ontology lool like??
<tmichel> RRSagent is out ...
<veroniqueM> Daniel: how to progress?
* fsasaki probably rrsagent has the flue
<veroniqueM> ... do we need a volunteer for this job?
<joakim> Yes, how do we define test for the ontology? Felix? Thierry?
<veroniqueM> ... any idea?
<raphael> Regarding test cases, raphael suggests to look at what other 
groups are doing
<veroniqueM> Daniel: Joakim suggested that we discuss this at the next 
face to face...
<raphael> ... there is also an attempt at the coordination group level (HCG)
<raphael> Media Fragments Test Cases: 
<veroniqueM> Thierry: as I said before, we should make a list of 
functionalities and assign some to some people in the group
<fsasaki> +1 to tmichel - this is probably rather simple, just somebody 
has to do the work
<raphael> Overview: 
* raphael zakim, unmute me
* Zakim raphael should no longer be muted
<veroniqueM> Daniel: I don't have any experience in W3C activities...
* florian zakim, mute me
* Zakim florian should now be muted
<veroniqueM> Raphael: all working groups have to build a test suite, so 
I suggest that you have a look at the other test suites/test cases
<pchampin> q+
* Zakim sees pchampin on the speaker queue
<veroniqueM> ... I have sent a link to a test suite/test case that we 
inspire ourselves from in the Media Fragment group
<joakim> +q
* Zakim sees pchampin, joakim on the speaker queue
<veroniqueM> Thierry: yes, as Raphael has said, there is not a single 
way to handle this, but in this group like in the Media Fragment start 
from scratch
<daniel> list of all funtionalities first...
<raphael> raphael is wondering what does it MEAN testing the ontology 
... it seems much easier for the API
<pchampin> zakim, unmute me
<Zakim> pchampin should no longer be muted
<veroniqueM> ... best way would be to list features that we have and 
make one test per functionality
<joakim> If we have to spec, we need to test suites
<raphael> q?
* Zakim sees pchampin, joakim on the speaker queue
<raphael> zakim, ack joakim
<Zakim> I see pchampin on the speaker queue
<raphael> zakim, ack pchampin
<Zakim> I see no one on the speaker queue
<veroniqueM> pa: like Joakim I don't know how we could test the ontology 
independently from the API
<joakim> s/to/two
<fsasaki> +1 to joakim and pa
<veroniqueM> + 1 from me too
<tobias> +1 from me as well
<veroniqueM> pa: the test suite would look like what the API 
would/should retrieve from a set of media resources
<veroniqueM> ... these resources will have different format/a specific 
<veroniqueM> ... should we have an implementation per format?
<raphael> By format, you meant video container / codec format?
<veroniqueM> ... an idea for a test suite is a set of resources with 
metedata in the format of the "formats in scope list"
<pchampin> yes
* raphael :-) thanks
<veroniqueM> pa: and check out for each property what it retrieves
<raphael> q+
* Zakim sees raphael on the speaker queue
<veroniqueM> Raphael: if the goal of the media ontology is to be 
compatible with existing formats, a testsuite would be to check how well 
the ontology is mapped to formats
<veroniqueM> pa: would this not be tested via the API, by seing what is 
retrieved across formats
<joakim> +1
<veroniqueM> Raphael: agrees, does not think of what it could mean to 
test the ontology independently
<tobias> @Veronique: but why do you need a test suite to check how well 
the ontology is mapped to the formats? you could do this by looking at 
them and arguing, imgo.
<joakim> We can group the tests to API- adn ontology related
<joakim> s/adn/and
<veroniqueM> ... but one part of/test suite could be to test the 
coverage of the mapping via the ontology
<veroniqueM> Daniel: we can have only one test suite document for the 
API and the ontology or do we need two separate documents?
<raphael> @tobias: you need a test suite, because you can have an 
implementation A that _claims_ to implement the Media Ontology, 
containing for example mappings between XMP and Media Ontology, but that 
cannot practically read XMP metadata
<veroniqueM> Joakim: @ Thierry, is it legal to group the two documents 
in one?
<veroniqueM> Thierry: it would be good to merge two documents (ontology 
and api documents)
* fsasaki q?
* Zakim sees raphael on the speaker queue
<raphael> zakim, ack me
<Zakim> I see no one on the speaker queue
<veroniqueM> ... and then we could have one test suite document
<veroniqueM> Daniel: there are several metadata standards mentioned as 
"in scope", what is the selection process for the test suite?
<raphael> The group should decide which metadata formats will actually 
be considered ... and provide in its documents the rationale for this choice
<raphael> The test suite should be exhaustive in the sense of testing 
all functionalities of the ontology/API
<veroniqueM> Tobias: we did this at the Barcelona F2F meeting based on 
the popularity of the schemata
<veroniqueM> Joakim: we had good ideas from pa, Thierry and Raphael for 
the test suite
<veroniqueM> ... we should dispatch the work
<raphael> I think the metadata formats to be considered by the test 
suite will be the ones mapped to the Media Ontology
<veroniqueM> pa: start with the schemas embedded in media resource, and 
list what is expected to be retrieved with the APII
<veroniqueM> Joakim: I could do it for XX
<veroniqueM> pa: I could do it for ID3
<joakim> Joakim can do EXIF
<veroniqueM> s/XX/EXIF
<veroniqueM> pa: I think we should use media resources with embedded 
metada, or create some...
<joakim> +q
* Zakim sees joakim on the speaker queue
<veroniqueM> and test how the API could access/read/retrieve from these 
resources' metadata
<veroniqueM> ... via the MAWG ontology
<raphael> PA: to answer your question, there are no media resources that 
embed MPEG-7 metadata, so there is nothing to test from your definition
<veroniqueM> Daniel: Florian collected info about requirements, but did 
not provide conclusions about selecting/reading metadata
<veroniqueM> Joakim: agrees with Raphael that we should not test EXIF 
metadata extraction
<veroniqueM> Raphael: the check should be independant of whether the 
metadata is embedded or not
<veroniqueM> pa: main challenge is to implement the mapping
<veroniqueM> Raphael: yes but 1st check that you can read/write MAWG 
<tmichel> Also FYI, Use Cases and Requirements for Ontology and API for 
Media Resource 1.0
<tmichel>  is a W3C WD 21 January 2010
<raphael> OK, I understand better what PA is saying, he is just one step 
ahead of me :-)
<tmichel> published at 
<veroniqueM> pa: I agree that extracting the metadata is one question, 
mapping the EXIF for ex to MAWG is another one
<raphael> So, I agree, test cases should test 1/ we can read/write 
ma:author metadata and 2/ we can test the mapping between ma:author and 
<veroniqueM> ... but there is EXIF metadata "outside" of resources
<raphael> @Joakim: there are plenty of libraries that extract EXIF metadata
<veroniqueM> ... so generating artificial metadata that way is problematic
<raphael> ... in all programming languages (JAVA, PHP, JS, etc.)
<raphael> ... most likely the same for ID3, XMP
<veroniqueM> Joakim: we could maybe recommend a library for extracting 
metadata from resources?
<Zakim> - +40.16.aacc
<veroniqueM> Daniel: pa can you send a mail summarizing the main points 
of this discussion?
<veroniqueM> Daniel: only 4-5 people at the F2F meeting in Korea
<pchampin> I'll send a mail
<joakim> Proposal: 7th F2F, Hosted by, Werner at JOANNEUM RESEARCH, in 
<joakim> After the MPEG meeting meeting 26-27 April
<raphael> 25-30 April conflicts with WWW conference
<veroniqueM> ... werner can host the following meeting in Vienna after 
MPEG7 meeting, april 26-30
<florian> +1 for F2F in vienna AFTER mpeg meeting
<joakim> Sorry proposed dates
<veroniqueM> ... who is in favor of this date?
<joakim> +1
<raphael> -1
<veroniqueM> Raphael: conflicts with www meeting
<tobias> +1
<daniel> +1
<wonsuk> +1
<veroniqueM> Daniel: can you show your opinions on irc, please?
<pchampin> not sure yet :-/
<veroniqueM> not sure yet either
* raphael thinks pa could have fun in the W4A conference :-)
<veroniqueM> Daniel: any other issue?
<veroniqueM> Raphael: RSS agent is not working, how can we get the 
minutes drafted?
* fsasaki somebody has to cut his local IRC log and send it to the list
<raphael> I volunteer to send right now plain text logs of the minutes
<veroniqueM> Raphael: I have an IRC log, can send the URL to list
<raphael> [meeting adjourned]
* florian zakim unmute me
<Zakim> -raphael
* fsasaki zakim, unmute Felix
* Zakim Felix should no longer be muted
* florian zakim, unmute me
* Zakim florian should no longer be muted
<Zakim> -tobias
<Zakim> -florian
<Zakim> -pchampin
<Zakim> -joakim
<Zakim> -Felix
* florian bye everybody
* Quits: florian (florian@ (Quit: Leaving)
* Quits: fsasaki (chatzilla@ (Quit: ChatZilla 0.9.86 
[Firefox 3.5.7/20091221164558])
<veroniqueM> Joakim: @Thierry we should check for next week so that it 
does not occur anymore
<veroniqueM> regrets: Chris
* daniel thanks raphael for your volunteering...
<Zakim> - +
* veroniqueM any other regrets?
<joakim> hej dŚ!

RaphaŽl Troncy
EURECOM, Multimedia Communications Department
2229, route des CrÍtes, 06560 Sophia Antipolis, France.
e-mail: raphael.troncy@eurecom.fr & raphael.troncy@gmail.com
Tel: +33 (0)4 - 9300 8242
Fax: +33 (0)4 - 9000 8200
Web: http://www.eurecom.fr/~troncy/
Received on Tuesday, 26 January 2010 13:08:13 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 16:24:38 UTC