MAWG -- 20 Jan 2009

<victor> Hello

<joakim> Hi Victor, all well?

<victor> fine, thanks

<joakim> who zakim, who is noisy

can you hear me ?

<joakim> scribe: Pierre_antoine

<raphael> scribenick: pchampin

<joakim> http://lists.w3.org/Archives/Public/public-media-annotation/2009Jan/0024.html

last minutes approved

<joakim> http://lists.w3.org/Archives/Public/public-media-annotation/2009Jan/0042.html

Frank will scribe for the next meeting

next call: January 27, 13:00 UTC unless it is unconvenient for the Apple people

<joakim> http://www.w3.org/2008/WebVideo/Annotations/track/actions/open

Action items

Werner: must have more discussion with Frank next week

action 47 still ongoing

<trackbot> Sorry, couldn't find user - 47

<raphael> Jean Pierre: try the web access to irc

<Zakim> raphael, you wanted to talk about the action from Felix

<raphael> http://labs.adobe.com/technologies/xmplibrary/

raphael: Felix and Raphael received mails from Adobe

about a library giving a better idea of the XMP properties actually in use

victor: worked on the Image UC

the UC is about image lifecylce

mailed a similar example involving video

<joakim> http://lists.w3.org/Archives/Public/public-media-annotation/2009Jan/0029.html

<joakim> Pierre Antoine: we should take structured data into account

there is a trade off between structured API and structured data

<raphael> Pierre Antoine, you should be able to watch all actions and issues, open and close ... so ask Thierry to get your access granted, this is a setup problem

close ACTION-59

<trackbot> ACTION-59 Investigate how the declaration of JSON could provide a solution for our return object closed

<raphael> http://www.w3.org/2008/WebVideo/Annotations/track/issues/open

joachim: manually adds an Issue: "trade-off btw structured API and structured data"

<raphael> the system should give a number to the issue

this is issue 1

<raphael> and you can assign actions related to this issue ...

close ACTION-65

<trackbot> ACTION-65 Finish TYoutube and Media RSS , TXfeed for next week closed

close ACTION-67

<trackbot> ACTION-67 Review tvanytime closed

close ACTION-68

<trackbot> ACTION-68 Review EBU closed

close ACTION-69

<trackbot> ACTION-69 Review Cable Labs Video-On-Demand Content specification 2.0 closed

close ACTION-70

<trackbot> ACTION-70 Review IPTC closed

close ACTION-72

<trackbot> ACTION-72 Review MIX closed

werner: 2 columns are missing in the mapping for SMPTE

they will be completed next week

close ACTION-75

<trackbot> ACTION-75 Review METS closed

close ACTION-76

<trackbot> ACTION-76 Review LOM closed

joakim: we can discuss on the mapping table

véronique: a lot of formats have a mapping to XMP and a mapping to DC

can we use the transitivity of the mappings ?

victor: not only the transitivity, but also symetry ?

<raphael> Does your mapping have a direction ? or are they equivalent relations in the mathematical sense?

joakim: in ID3, a property maps to both XMP and DC, but not exactly in the same way

DC is more general, XMP more specific

werner: same observation: there is often an overlap

<raphael> the mapping can be precised: sub-property, equivalent-property, related-property, etc. OWL and SKOS provide constructs to express that

many properties are music-oriented; should we keep them?.

véronique: the mapping of MPEG7 will probably provide properties related to video

that we don't have yet: mapping problem

<scribe> ACTION: joakim to merge the contributions to the mapping table on the wiki [recorded in http://www.w3.org/2009/01/20-mediaann-minutes.html#action01]

<trackbot> Created ACTION-77 - Merge the contributions to the mapping table on the wiki [on Joakim Söderberg - due 2009-01-27].

<VeroniqueM> for VRA at least they provided a clear mapping to Dublin Core 4.0

<VeroniqueM> and to former versions, but i did not include them...

pierre-antoine: is the mapping of DC vs. DCterms always consistent with the 2 versions (string / URI) provided by the Dublin Core?

<scribe> ACTION: pchampin to start a discussion on the wiki about the DC vs. DCterms issue [recorded in http://www.w3.org/2009/01/20-mediaann-minutes.html#action02]

<trackbot> Created ACTION-78 - Start a discussion on the wiki about the DC vs. DCterms issue [on Pierre-Antoine Champin - due 2009-01-27].

<joakim> http://www.w3.org/2008/WebVideo/Annotations/wiki/Dataformat

data format

<raphael> I think that URI as values is a must ... to be part of the Web of Data

<raphael> ... this is coherent with the vision of a Web of things

<VeroniqueM> I agree with Raphael

<wbailer> +1

<joakim> +1

<victor> +1

<wonsuk> +1

<joakim> http://lists.w3.org/Archives/Public/public-media-annotation/2009Jan/0032.html

<VeroniqueM> one way of representing the mappings could be in a Semantic Web format, what are the other formats that people would want it into?

<VeroniqueM> with skos:mappings for example

véronique: SKOS is about specifying different kinds of mapping (exact mapping, more general, etc)

<VeroniqueM> http://www.w3.org/2004/02/skos/

<victor> i absolutelty agree about SKOS

<scribe> ACTION: VeroniqueM to write an example about using SKOS [recorded in http://www.w3.org/2009/01/20-mediaann-minutes.html#action03]

<trackbot> Sorry, couldn't find user - VeroniqueM

<raphael> trackbot, status

<scribe> ACTION: Véronique to write an example about using SKOS [recorded in http://www.w3.org/2009/01/20-mediaann-minutes.html#action04]

<trackbot> Created ACTION-79 - Write an example about using SKOS [on Véronique Malaisé - due 2009-01-27].

<VeroniqueM> there are two points: the API and the media ontology...

pierre-antoine: do we expect people using our API to be able to deal with SKOS ?

<VeroniqueM> they are related, of course, we agree with that

<Zakim> raphael, you wanted to explain a bit more how SKOS can be useful

raphael: SKOS is for representing lightweight taxonomies

without strong semantic comitment

joakim: should we include MicrosoftTV on iTunes ?

(no opinion expressed)

- DRAFT -

MAWG

20 Jan 2009

Attendees

Contents

Action items

data format

Summary of Action Items

Scribe.perl diagnostic output