This section is normative; however, examples contained in this section are informative.

The following table lists the formats that were selected as in-scope of a potential mapping from the Media Ontology, along with the identifiers which are used as prefixes to identify them in this specification.

We distinguish multimedia metadata formats that focus on the description of multimedia resources from multimedia container formats. In the case of the latter, only few technical properties are relevant for the Ontology for Media Resources, because of they widespread usage. Very specific properties are out of the scope of this specification

The following formats are out of scope for this specification.

"A Uniform Resource Identifier", or URI, is defined in []. In this specification, the term URI is used, since it is well known. However, the use of this term is extended in this specification to also include "Internationalized Resource Identifiers" (IRIs), as defined in []. An IRI is a URI that MAY contain non-escaped characters other than ASCII characters. The data type is anyURI. Hence, in this specification, the term "URI" MUST be interpreted to also include IRI.

A String value MUST be represented using the XML Schema string data type.

An Integer value MUST be represented using the XML Schema integer data type.

A Decimal value SHOULD be represented using the XML Schema decimal data type, but MAY be represented using the XML Schema double data type if decimal is not available.

A Date value MUST be represented using one of the specific date/time data types of XML Schema, depending on the available precision: gYear gYearMonth, date, dateTime, or dateTimeStamp.

Concerning external subtitles, using relation is the recommended approach. The identifier attribute contains the URL of the subtitle file, and the relation type qualifies it as a subtitle relation. The value should be a URI, but could also be a string. It is recommended to use a controlled vocabulary for the type of the relation.

Embedding of subtitles is not a use case that has been considered, however it is possible. The mechanism used to specify timed metadata is to specify fragments identified by Media Fragment URIs [] and then describe annotations of these fragments.

To summarize, there are three options for dealing with subtitles:

Although the last option is a way of embedding subtitles it is not recommended. Instead, a dedicated format such as TTML or WebSRT should be used for the subtitles and referenced.

Time based annotations are a possible and the following two cases are covered by the specification:

At the time of writing this specification, there no solution for embedding a set of triples into one of the properties of the Ontology for Media Resources 1.0. The summary of a discussion with the Semantic Web Coordination Group is that named graphs could be a solution to this issue, but there is no standard syntax for expressing them, to which this specification could refer. Such a syntax might find its way into RDF 2.0. Thus the embedding of triples into media annotation elements is excluded until a standard syntax for named graphs is available.

Core property definitions section defines a gerenal property fragment with a role attribute to specify the relation between the resource and its fragment, like captioning or signing. In the RDF representation, this is achieved by defining subproperties of the <tt>ma:hasFragment</tt> property.

Captions and signing of a media resource can be provided in different forms, the most typical being:

To account for this diversity, the RDF ontology does not link <tt>ma:hasTrack</tt> with <tt>ma:hasCaptioning</tt> or <tt>ma:hasSigning</tt>. The last two can link a media resource to any fragment, e.g. a spatial fragment of the video track where the signing is located, or even an external file considered as a fragment of the resource. If the fragment is also a track, nothing prevents to link it with both properties <tt>ma:hasCaptioning</tt> and <tt>ma:hasTrack</tt>.

For example, the following RDF describes a video with embeded signing, subtitles as an external file, and a track containing audio-description (caption for accessibility):

  <video.ogv> a ma:MediaResource ;
    ma:hasSigning <video.ogv#xywh=percent:70,70,90,90>;
    ma:hasSubtitling <./video.srt> ;
    ma:hasAudioDescription <video.ogv?track=subtitle> ;
    ma:hasTrack <video.ogv?track=subtitle> ;

The core set of properties proposed in section 5 only defines a single property for specifying the language of a media resource. However, a media resource may have several languages. For example, a video file can have the following languages applying to it:

The four language codes could be directly applied to the video file, using the language core property <tt>ma:hasLanguage</tt> in the RDF representation), but this would lose a part of the information.

If one wants to keep the complete information, the recommended option is to assign each language to the appropriate fragment of the video, using [MediaFragment] to identify them, and the core property fragment <tt>ma:hasFragment</tt> and its subproperties in the RDF representation to link them to the video file itself. In the example above, we would have:

The corresponding RDF would be:

  <video.ogv> a ma:MediaResource ;
    ma:hasTrack <video.ogv#track=audio>,  
                <video.ogv#track=subtitle>;
    ma:hasSubtitling <video.ogv#track=subtitle> ;
    ma:hasSigning <video.ogv#xywh=percent:70,70,90,90>  .  

  <video.ogv#track=audio> a ma:AudioTrack ;
    ma:hasLanguage [ rdfs:label "en-GB" ] ;
    ma:hasFragment <video.ogv#track=audio&t=10,20> .

  <video.ogv#track=audio&t=10,20>  a ma:MediaFragment ;
    ma:hasLanguage [ rdfs:label "fr" ] .

  <video.ogv#track=subtitle> a ma:DataTrack ;
    ma:hasLanguage [ rdfs:label "es" ] .

 <video.ogv#xywh=percent:70,70,90,90>  a ma:MediaFragment ;
    ma:hasLanguage [ rdfs:label "bfi" ] .

The namespace of the Ontology for Media Resources 1.0 is defined by this URI: http://www.w3.org/ns/ma-ont#. Applications that are compliant with this specification MUST use this namespace URI.

The ma prefix name is associated with the namespace URI http://www.w3.org/ns/ma-ont# in this document.

The following table gives the correspondence between the core properties as described in the Descriptive properties (Core Set)section and the RDF vocabulary given below.

Unless stated otherwise, atomic values are represented by literals while complex values are represented by resources. It follows that, in the general case, properties with complex values are represented by object properties, while properties with simple values are represented by datatype properties. Attributes in complex values are represented by properties of the resource representing the complex value; depending on their semantics, they are represented by datatype or object properties.

The RDF ontology also introduces a number of classes corresponding to the domains and ranges of the corresponding property.

The following is the authoritative RDF/OWL representation of the Media Ontology: the Ontology for Media Resources 1.0

 

       &ma-ontology-owl;