<daniel> talking about reviewers on our documents: asked to webapps wg and looking for someone.

<daniel> werner, good morning, are you supposed to participate in teleconf now ?

<daniel> we will have a document review (ontology) for two hours from now. will you join ?

<mediaann> up and running

<daniel> we start meeting

<daniel> starting meeting and self-introduction

<daniel> starting meeting and self-introduction

<daniel> Daniel, Hui, Mingyo, Chris, Thierry, Joakim, Wonsuk, and Werner (online)

<daniel> Werner, it's your turn, please introduce yourself

<chris> scribe: Chris

<scribe> Meeting: MAWG F2F Korea

<scribe> ACTION: Wonsuk to request review from MXM during last working draft call [recorded in http://www.w3.org/2010/02/23-mediaann-minutes.html#action01]

<trackbot> Created ACTION-214 - Request review from MXM during last working draft call [on WonSuk Lee - due 2010-03-02].

Review and update of Ontology for Media Resource 1.0

<daniel> Agenda bashing: http://www.w3.org/2008/WebVideo/Annotations/wiki/Meeting_Agenda_Seoulf2f

<daniel> Starting ontology document review and update

wonsuk: Veronique sent an email on the update of the ontology doc
... had problems to validate

<jsderber> http://winmerge.org/

http://validator.w3.org/

we are currently looking into validating the document of Veronique

<cpoppe> scribe: cpoppe

we fixed the file from Veronique

werner: we should give a clear picture of what the document will look like
... important is the type issue

joakim: I am not sure we will get type definitions from all the formats

werner: I agree, maybe we should use the ones we have
... we should define the types for the core properties based on these

wonsuk: we could make a draft based on the current mapping table

what format should we use for the datatype properties

werner: the datatypes of the properties should be added in section 3

joakim: first issue; work on datatypes
... should we work on candidate additional elements

http://www.w3.org/2008/WebVideo/Annotations/wiki/Candidate_Additional_Elements

werner: maybe we should wait with this for now

<jsderber> Descriptive metadata elements

<jsderber> Accessibility metadata

<jsderber> Discovery of tracks

<jsderber> Resolved Issues

<jsderber> Accessibility metadata

<jsderber> comment from PFWG [2]: request elements for referencing alternate version

joakim: ok first thing; work on the datatypes

werner: would be good to add the properties in the document to get feedback
... maybe a seperate section

thierry: we should look at our use cases and requirements doc
... if the properties are not needed why should we add them?
... it's better to add stuff now and remove it afterwards if there is no interes

werner: the properties requires redefining ma:tracks
... this introduces an API issue to access the tracks

<wbailer> http://www.w3.org/WAI/PF/HTML/wiki/Media_MultitrackAPI

werner: we could change the definition of ma:fragments for track fragments

<tmichel> projecting the last version of the Ontology document

datatypes

<tmichel> wonsuk will report tomorrow on the last chances of the mapping table

<tmichel> datatype of core properties ...

<jsderber> http://lists.w3.org/Archives/Public/public-media-annotation/2010Feb/0047.html

<tmichel> looking at a email from Werner about the Ontology in the archives

<tmichel> Let's check if these comments are updated into the spec

<tmichel> in abstract replace "common" with "core"

<tmichel> - abstract: "interoperability among various kinds of metadata formats" suggests that we would also support converting between formats A and B

<tmichel> not resolved TDB.

<tmichel> - intro: "describe mapping between media resources" is not what we are doing

<tmichel> we are not mappting between media resources but between properties (metadata)

<tmichel> - intro, 3rd bullet about DC: "we might want to apply other restrictions" - do we want to be more specific here and say that we want to define stricter semantics, e.g. ma:format (MIME type of the resource) vs. dc:format (The file format, physical medium, or dimensions of the resource.)

<tmichel> add example suggested by Werner

<tmichel> resolution: add example suggested by Werner

<tmichel> - intro, last paragraph: we still say that the set of elements is "likely to be extended", I think we should consider now whether we want to add any of the elements discussed in [2] and drop this paragraph before going to LC

<tmichel> we will drop it. and add other candidate properties. Will add media tracks.

<tmichel> to ask for feedback

<tmichel> "likely to be extended" will be removed

<tmichel> wiil add the Discovery traks in the candidate list of prerties

<tmichel> - 1.3: we should give some explanation why MPEG-21 is not in scope (e.g. not media description in narrower sense)

<tmichel> not really about describing ressources ... not media description but description of the media chain

<tmichel> MPEG 21 deals with how the media will be consumed. Not media description.

<tmichel> - def. media resource: add links to FRBR, BBC ontology

<tmichel> OK will add

<tmichel> - def. property: the text talks about string or URI as value, but I think we should explicitly say that values of properties can be structured and/or unstructured

<tmichel> Make it clear that the property can be structured and/or unstructured

<tmichel> remove in the example th "simple string or as the URI ..."

<tmichel> - references: B has heading "References (Non-Normative)", so I assume A should be "References (Normative)"; however, I'm not sure if all the references are really normative, e.g. HTML5 (not referenced), MPEG-21 (out of scope)

<tmichel> should have a subsection for normative vs non normative references. For example MPEG21 is informative as for HTML5 when EXIF is normative

<tmichel> - references: the FRBR reference is incomplete

<tmichel> - references: the MPEG-21 reference is also incomplete - but nonetheless has a typo ;-)

<tmichel> To be fixed

<tmichel> Statement about the "Members of the Group present at the F2F meeting in Barcelone" should be changed

<tmichel> the resolution is a WG resolution if not objected.

atatypes

datatypes

<tmichel> http://www.w3.org/2008/WebVideo/Annotations/drafts/ontology10/WD/summary.html

<tmichel> we can also look at the types from the API document

<tmichel> for example identifier is a "DOMsrting" as a return type

<tmichel> could use it for the type

<tmichel> ma:identifier type is a URI

<tmichel> ma:title string

<tmichel> ma:title is a string

<tmichel> ma:language is a string (BCP47)

<tmichel> ACTION: BCP47 obsoletes RFC4646 ...should be change in the mapping table and Ontology document [recorded in http://www.w3.org/2010/02/23-mediaann-minutes.html#action02]

<trackbot> Sorry, couldn't find user - BCP47

<tmichel> ACTION: wonsuk BCP47 obsoletes RFC4646 ...should be change in the mapping table and Ontology document [recorded in http://www.w3.org/2010/02/23-mediaann-minutes.html#action03]

<tmichel> ma:locator is URI

<tmichel> ma:contributor is a pair (identifier and role)

<tmichel> it is a complex object. We are not looking for a return type here.

<tmichel> the API does specify the return type

<tmichel> could describe with something like {String or URI, String}

<tmichel> or smeting {Identifier:String or URI, subproperty:String}

<tmichel> let's look at JSON ...

<chris> http://json-schema.org/

<wonsuk> datatype of contributor --> { Identifier:URI, role:String }

<wonsuk> datatype of contributor --> { Identifier:URI }

<wonsuk> datatype of contributor --> URI

<tmichel> ma:creator is a URI

<wonsuk> datatype of creator --> URI

<tmichel> ma:createDate is a date

<tmichel> ma:location is currently undefined in the API ... like a city a GPS an address ...

<tmichel> could be a string

<daniel> http://www.ietf.org/rfc/rfc5139.txt for geolocation reference

<tmichel> [Name:String], [longitude:Float,latitude:Float, alitude:Float]

<tmichel> [Name:String], [longitude:Float,latitude:Float, alitude:Float], [Address:String]

<tmichel> let's keep it simple. If one needs the address the link will enable quering the original format. For example if the is "more general"

<tmichel> resolution for ma:location is [Name:String], [longitude:Float,latitude:Float, alitude:Float]

<tmichel> ma:location is a string

<tmichel> ma:keyword is a string or a set of strings

<tmichel> ... more a bag of strings...

<tmichel> description of ma:keyword a set of descriptive phrases

<tmichel> for the datatype field ma:keyword is a string ?

<tmichel> resolution: description of ma:keyword "descriptive phrase or keyword"

<tmichel> ma:keyword is a string

<tmichel> ma:genre is a string

<tmichel> more complex than a String ma:rating need to know the person or organization and the value of the rating

<tmichel> ACTION: Daniel to remind David Singer about pending actions [recorded in http://www.w3.org/2010/02/23-mediaann-minutes.html#action04]

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

<tmichel> [Identifier:URI, value:Float,max:Float,min:Float,context:String]

<tmichel> ma:relation

<tmichel> [Identifier:URI, relation:String]

<tmichel> ma:collection is string

<tmichel> ma:collection is a complex object

<tmichel> [copyright:String, Identifier:URI] for ma: copyright

<tmichel> ma:licence

<tmichel> [licence:String, Identifier:URI]

<tmichel> ma:publisher

<tmichel> is a URI

<tmichel> ma:targuet Audience

<tmichel> [Identifier:URI, classification:String]

<tmichel> ma:fragments

<chris> hi Werner, we are looking at the ma:fragments property

<chris> any suggestions on how you would change the definition?

<jsderber> Link to email: http://lists.w3.org/Archives/Public/public-media-annotation/2010Feb/0051.html

<tmichel> will review ma:fragments later.

<wbailer> A list of pairs of fragment role and fragment identifier. Fragment types can be spatial, temporal or tracks (e.g. chapters, favourite scenes, subtitles).

<tmichel> ma:frameSize

<tmichel> [Width:Float, Height:Float]

<tmichel> ma:compression

<tmichel> is a string

<tmichel> ma:duration is a float

<wbailer> should we consider when a string is from a controlled set of terms (e.g. for ma:compression)?

<tmichel> ma: format is a string

<tmichel> ma:samplingrate is a float

<tmichel> ma:framerate is a float

<tmichel> ma:biterate is a float

<tmichel> ma:numTracks is an integer

<chris> werner you mean that we enforce this string should be from the controlled set of terms?

<wbailer> not sure if "enforce", but typically the value will be one from a set

<wbailer> i think we discussed that earlier, that we want to support controlled terms for several properties, but in addition leave the option for free text

<chris> yes free text is allways available through the unstructured return value in the API

<chris> what do you mean with "support" controlled termes

<chris> terms

<wbailer> stating for which properties there will be such a set of terms, and reference an appropriate one (e.g. MIME Types, EBU or MPEG classification schemes)

<chris> yes in the description of the property an existing set of terms is recommended

<wbailer> ok

<chris> but that's how it already is now, you suggest we make this a restriction?

<chris> meaning that we need to specify the mapping to this controlled set?

<wbailer> i'm not sure if it like that now - eg in the property definitions of the summary table we make no mention of controlled terms

<wbailer> as we leave the option for free text, i do not think we need to define mappings of the terms

<chris> for compression we refer to RFC 4281, for format we refer to MIME type

<wbailer> it says "possible use" for the compression, i think we could be a bit more strict

<wbailer> ... in the sense: one can use free text or a controlled term

<wbailer> ... if it's a controlled term, it's from a specified set

<tmichel> ACTION: MAWG members please review the datatype and raise issues by tomorrow morning if needed. [recorded in http://www.w3.org/2010/02/23-mediaann-minutes.html#action05]

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

This is scribe.perl Revision: 1.135  of Date: 2009/03/02 03:52:20  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/contorlled/controlled/
Succeeded: s/specifed/specified/
Found Scribe: Chris
Inferring ScribeNick: chris
Found Scribe: cpoppe
Inferring ScribeNick: cpoppe
Scribes: Chris, cpoppe
ScribeNicks: chris, cpoppe
Default Present: wbailer, +82.10.58.00.aaaa
Present: wbailer +82.10.58.00.aaaa

WARNING: Fewer than 3 people found for Present list!


WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Got date from IRC log name: 23 Feb 2010
Guessing minutes URL: http://www.w3.org/2010/02/23-mediaann-minutes.html
People with action items: bcp47 daniel mawg wonsuk

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.