Improvements / Enhancement for a revision of the "Ontology for Media Resource 1.0" document

From Media Annotations Working Group Wiki

Genereal information:

This section accumulates possible improvements or enhancements for a revision of the "Ontology for Media Resource 1.0" document. These items came up in telephone conferences and the public mailing list of the MAWG. The collection´s starting point is April 2009 (till 12.06.2009). Some items of this collection might be allready closed - please add the appropriate state.

List of abbreviations:
n.y. - not yet
d. - done
c. - canceled
o.i. - open issue
m.d. - metadata

Accumulated topics:

Nr.	State	Date	Topic	Description	Proponent	Link to initial source (mailing list or meeting minutes)
13	ongoing	05.06.2009	Data type description in the ontology doc	The datatypes of the properties should be choosen in accordance to rhe standards in scope.	W. Bailer	http://lists.w3.org/Archives/Public/public-media-annotation/2009Jun/0013.html
15	ongoing	09.06.2009	Section 3.2	refers to item 13	J. Söderberg	http://www.w3.org/2009/06/09-mediaann-minutes.html
16	ongoing	09.06.2009	Missing: Syntactic and semantic mapping	Syntactic mapping e.g., mapping between name and firstname surname, refers to 13	F. Sasaki	http://www.w3.org/2009/06/09-mediaann-minutes.html
18	ongoing	09.06.2009	Layers of conformance	It should be possible to create different "layers of conformance" in the ontology, representing different degrees of formality.	W. Bailer	http://www.w3.org/2009/06/09-mediaann-minutes.html
19	ongoing	09.06.2009	Defining types for properties	For elements that contain type/value pairs, such as contributor, etc. EBU classification schemes will be used as a starting point	S. D. Park	http://lists.w3.org/Archives/Public/public-media-annotation/2009Jun/0027.html

Ongoing items:

Nr.	State	Date	Topic	Description	Proponent	Link to initial source (mailing list or meeting minutes)	Volunteer
6	ongoing	19.05.2009	Mapping table vs. formats in scope	A few "in-scope" formats are not part of the mapping table. (Don´t know if this is allready sufficiently done in the beginning of section 4.2.2)	V. Malaise	http://www.w3.org/2009/05/19-mediaann-minutes.html	Veronique
7	ongoing	20.05.2009	Improvement of tag descriptions	There were some efforts in improving the descriptions of the MAWG Core Properties. Allready finalized?	D. Singer & W. Lee	http://lists.w3.org/Archives/Public/public-media-annotation/2009May/0099.html & http://www.w3.org/2009/05/26-mediaann-minutes.html	Veronique and David
8	ongoing	20.05.2009	Clarification of Sec. 4.1.1	This section is supposed to be enlarged by some further information	W. Lee	http://www.w3.org/2009/05/26-mediaann-minutes.html	Wonsuk
12	ongoing	02.06.2009	Adding the types	Adding types to the core properties should be part of the next draft. F2F St. Clara: Reordering of the properties and grouping should be in scope (see summary table). Editors should review their part in the mapping table (see MPEG-7)	S. D. Park	http://www.w3.org/2009/06/02-mediaann-minutes.html	Daniel (coordination)
11	ongoing	02.06.2009	Issue "ontology"	we never define ontology, and never state that we consider a light weight description of properties and their informal mappings as an ontology.	V. Malaise	http://lists.w3.org/Archives/Public/public-media-annotation/2009Jun/0004.html	Veronique
14	n.y.	05.06.2009	Loss of semantics	Some mappings will produce a loss of semantic information (actor <-> pincipal protagonist). This issue should be adressed in the document.	D. Singer	http://lists.w3.org/Archives/Public/public-media-annotation/2009Jun/0016.html	Veronique

Closed items:

Nr.	State	Closing Date	Topic	Description	Proponent	Link to initial source (mailing list or meeting minutes)
10	n.y.	02.06.2009	Renaming of section 4	Section 4 should be renamed, we do not define the properties, we simply list the properties taken into account in the core part of the ontology.	V. Malaise	http://lists.w3.org/Archives/Public/public-media-annotation/2009Jun/0004.html
1	n.y.	25.06.2009 (F2F)	Browseable version of the ontology	A browseable (wiktionary-like) version of the proposed ontology could improve the understandability for people, who are not familiar with it.	R. Tous	http://lists.w3.org/Archives/Public/public-media-annotation/2009Apr/0072.html
2	d.	25.06.2009 (F2F)	Adding porperties regarding descriptive and technical m.d.	There are a few properties, which could possibly improve the expressivness of the proposed ontology: tag line, reference, sample type/depth, file size and temporal sampling rate.	W. Bailer	http://lists.w3.org/Archives/Public/public-media-annotation/2009Apr/0085.html & http://www.w3.org/2008/WebVideo/Annotations/wiki/Candidate_Additional_Elements
3	d.	25.06.2009 (F2F)	Adding porperties regarding rights management	Rights management is a crucial aspect of any media consumption task. The group should not dig into the complexity of rights expressions, but perhaps add some simple property.	T. Bürger	http://lists.w3.org/Archives/Public/public-media-annotation/2009May/0000.html & http://www.w3.org/2008/WebVideo/Annotations/wiki/Candidate_Additional_Elements
4	d.	25.06.2009 (F2F)	Adding porperties for types of fragments	The two properties spatialMediaFragment and temporalMediaFragment may improve the expressivness of the ontology.	F. Sasaki	http://lists.w3.org/Archives/Public/public-media-annotation/2009May/0001.html & http://www.w3.org/2008/WebVideo/Annotations/wiki/Candidate_Additional_Elements
5	c.	25.06.2009 (F2F)	ISAN	Should it be in-scope or out of scope? Is it relevant?	S. D. Park	http://lists.w3.org/Archives/Public/public-media-annotation/2009May/0059.html
9	n.y.	29.06.2009	Unclear terminology	What´s the exact meaning of "semantics-preserving"? Maybe rephrase? (Werner also suggests to drop it, Felix proposed to replace it by "semantic level mappings")	V. Malaise	http://lists.w3.org/Archives/Public/public-media-annotation/2009Jun/0004.html

Why not use DC (names space)?

or

Why copying of the Dublin Core elements into a different namespace?

MAWG Answers/motivation:

1) The group sees the ma: properties – not as replacements for the DC and other properties, but as identifiers for mappings between equivalent properties in various formats.

2) (Felix) There seems to be a misunderstanding in this thread about what "ma:" is. These are not new properties meant for usage themselves, but means to interrelated existing properties. The main outcome of the group is the mapping of "ma:" to other formats, that is the ontology. Currently this is a table, in the future the mapping might be represented differently.

Another outcome will be an API to access metadata, but again not for reading or writing "ma:" properties directly, but rather the scenario is: you ask for "ma:creator", but will get "dc:creator" or whatever is in the media file, and you ask for writing a "creator", but we will not write "ma:creator" in the media file, but what is appropriate for the format.

3)

Another might be that DC is simply one of 25+ schemes investigated, all of which have some common aspects. Why bless DC rather than the others? A DC response here would be that Dublin Core is intended to be a lightweight common core, created for just such purposes, ie. mixing information across domains.

4) Note also that the API we are working on, see http://dev.w3.org/2008/video/mediaann/mediaont-api-1.0/mediaont-api-1.0.html is even more specific than dublin core, e.g. in terms of return types.

General comment:

Exposing metadata is tricky, often giving a choice between creating an endless ontology or building an open-ended system that guarantees no interoperability. " I am not sure what you are looking for: - an ontology defining mappings between existing metadata (being defined by media annotations working group (MAWG)) - new metadata properties defined as an ontology (not defined by MAWG) - an API which reflects the mapping of existing properties in the ontology to access methods for metadata (provided by MAWG) - an API for low-level reading mechanisms (not provided by MAWG - e.g. for XML-based formats provided by an XML-parser)