SUDS is Used to Describe Soaps (SUDS) is a vocabulary for describing the world of soap opera characters. Its name, a recursive acronym, stands for ‘SUDS is Used to Describe Soaps’. It is an application of Resource Description Framework (RDF) to a rich set of relationships and content as found in the world of the soap opera. The model is largely concerned with characters relationships to each other, which is the area of most interest to the programme’s audience. Other aspects of the model describe events in the programme and the world outside the fictional sphere of the programme e.g. awards that actors had won, and other programmes actors had appeared in.

The vocabulary was developed as a result of a project to redesign the EastEnders web site. While it is based on the material found in EastEnders it has been designed so that it would be able to be used for other programmes of the same genre.

EastEnders is an enormously popular, long-running British TV soap, which is set in the East End of London and produced by the British Broadcasting Corporation (BBC). It is a gritty, realistic character-driven programme that focuses on a multicultural working class community. It is watched by 14 million people, approximately a quarter of the UK population, and has an accompanying web site which is correspondingly popular. By 2002, however, the site was in great need of a redesign and a more effective content management system.

The BBC’s Content Management Culture (Content Management Culture (CMC)) group was called in to help with the project, which had two main goals: firstly, to improve navigation and update the site’s visual design, and secondly, to migrate to a content management . The CMC’s primary role was to define content objects, create content models and describe metadata schemes, while advising on content management policy and creating controlled vocabularies. However, we also saw the opportunity to model the relationships within the soap opera in a way which augment the more conventional content management tools that were already planned.

The inspiration for the SUDS vocabulary came from our knowledge of the FOAF vocabulary. FOAF, which was originated by Dan Brickley and Libby Miller, stands for ‘Friend of a Friend’, and is a way of describing people and their relationships with each other using XML/RDF.

SUDS is an extension of the FOAF vocabulary. FOAF currently describes the relationships between people by saying that a person ‘knows’ another person e.g. Barry knows Paul; Paul knows Sam. As we can assume that, in general, everyone in a soap opera knows each other, we can take this simple relationship for granted. Instead, we describe their relationships through events e.g. Laura is married to Ian, therefore they are husband and wife. Or in another, more soap opera-like example: Ian is having a secret affair and shot his lover’s husband.

Currently, SUDS contains several classes that are not found in FOAF:

We also added a number of properties, many of which allow us to define types of events, so that we can describe the events that occur in soap operas (marriages, affairs, arrivals, disappearances, inheritances, murders, etc).

The primary gain that we recognised would result from relationship modelling was easing the team’s tasks of day-to-day content creation and maintenance.

The web site team is made up of five people, who keep the site’s content up to date and commission development work. They spend a significant amount of time creating Episode Updates, which are blow-by-blow accounts of what has happened in each episode. People who miss an episode can come to the site, read the Update, and be no longer be ‘behind’ with what’s going on in the EastEnders world. Other content creation tasks undertaken by the team include creating character and actor profiles when there is a new character, and maintaining them as necessary to ensure they are up to date. The team also answers users’ questions about characters and the programme, and write quizzes, mini-sites and other features.

Currently, the team has three resources for content creation:

Compare this with the character information reference system of The Archers, which is a serial that has run for over forty years on BBC radio. There, character information is kept on a system of over 20,000 index cards.

With SUDS, we can take some of the work out of this process.

To develop SUDS, we needed to understand the scope of the content domain and the kinds of events and relationships within it. We used several methods to reach this understanding. We:

We also researched the area of ontologies, and came across the (temporary) pitfall of trying to follow an existing model too closely. For example, one referenced paper [07]describes an ontology for wine, but soap operas did not fit easily into this model. In a model for wine, we are interested in an individual wine or a group of wines, whereas in soap operas we are interested primarily in the relationship of one thing to another. The situation was resolved by returning to the EastEnders content and the questions that were asked of it, so that a more appropriate model could be developed.

Through this we came to our content model, where characters are related to actors and locations; actors are related to programmes and awards; and characters are linked to episodes and storylines.

What constitutes a soap storyline?

A traditional storyline is often defined by what Christopher Vogler, in his book ‘The Writer’s Journey’. Put simply, there is a central character who is introduced in their ‘ordinary world’, embarks on a journey of some kind, and ultimately returns home, transformed by their adventure. This pattern can be seen in stories from The Odyssey, to the Wizard of Oz, to Star Wars. It is also evident in television series: in some sophisticated examples, this structure can be seen at the levels of episode, season and programme.

In contrast, what we see in soap operas is what cultural critics refer to as an ‘open text’: a narrative containing multiple overlapping storylines, with a multiplicity of voices, and a large amount of indeterminacy governing the resolution of any one thread. Much of the action in a soap opera is taken up by the conversations of the characters discussing the implications of events in their lives rather than a development of a series of connected, interlocking events that achieve resolution. Realising this, we removed the storyline object from our early model. However, if we designed the model without any refernce to this aspects of the soap opera's content, we would be left with a rather dull enumeration of events such as weddings, births and divorces.

We know from audience studies that what people find most enjoyable about soap operas are the ‘in between’ discussions and negotiations that occur between characters (see Buckingham’s study[05]). The way that we resolve these issues in the SUDS model is to link out from events to Episode Updates, so that the SUDS interface becomes, in effect, a way of navigating the programme episodes via events and characters rather than the simple chronology that is presented currently.

The next steps for SUDS development will involve further development of the vocabulary through tests, iteration and deployment. There is potential to use the vocabulary for both the EastEnders web site and other sites across the BBC. Already, other programmes of a similar genre, for example the BBC’s radio series ‘The Archers’, have expressed interest in using the SUDS vocabulary to model their own content.

In addition, the more actor/programme focused elements of the model could be used to share content about actors, programmes and characters across the BBC. Content sharing has already been identified as a project that would bring benefits to the BBC, enriching content and reducing the duplication of work.