New Media: a narrative approach to content annotation

Recent years have seen an upsurge in the popularity of user-generated content. Sites such as Youtube1 and Flickr2 have illustrated that increasing numbers of web users are willing to publicly share their content, while equally sites such as Delicious3 and Blinklist4 demonstrate that growing numbers of users are willing to annotate each other’s content. Annotation, in this context, comes primarily under the guise of social tagging whereby users apply labels to resources in a subjective yet non-restrictive approach to subject-based indexing. The popularity of tagging has given rise to the folksonomy5 [1, 2] as an alternative to the less community-centred taxonomy. However, there are limitations to the use of the folksonomy. Tagging a resource with meaningless data will not necessarily lend a resource meaning [3]. Moreover, there has been an industry wide push to develop metadata schemas, vocabularies and ontologies to help standardise the way in which audiovisual information is archived, accessed, retrieved, filtered and managed. From this perspective, while MPEG-7 [4] introduced a comprehensive metadata schema to help organise media content, IMDI [5] proposed a metadata standard to help describe multimedia and multimodal resources. There is further work being carried out with ontologies, such as the DOLCE ontology [6] for Linguistic and Cognitive Engineering and COMM [7], the Core ontology for the description of multimedia assets. Although more descriptive vocabularies present the possibility of repurposing media content, they are not widely adopted and when they are it is often in an experimental capacity. This reintroduces the question of annotation. It could be argued that the more expressive a vocabulary the greater the burden on the user to annotate a resource in its entirety. This problem is exacerbated when presented with user-generated content. It is unrealistic to expect the non-professional end-user to undergo prolonged periods of annotation when wishing to submit media content. In this paper we consider a narrative approach to annotation that will help to bridge the gap between low-level subject-based annotation, as evinced with social tagging, and more expressive metadata vocabularies, such as MPEG-7. The approach is predicated on developing a narrative authoring environment where the user develops beats, events, characters, plotlines and finally stories. Each story is buttressed by user-generated content. However, we suggest that this occurs within a specific narrative context, for example, to illustrate a plotline or as piece of dialogue between two characters. Therefore each piece of user-generated content is automatically provided with overarching metadata supplied through reasoning with the narrative context. The approach seeks to remove the burden of annotation from the end-user yet automatically describe resources to a high level of detail with expressive and standardised vocabularies.