Information Technology Reference
In-Depth Information
“hooks” on the right level of abstraction: fine-grained enough to allow signif-
icant combinatorial possibilities and the capability for surprise, yet providing
the appropriate authorial affordances to allow the exertion of authorial control
over multiple levels of the story construction process.
The
Te rmi na l Time
architecture
Te rmi na l Time
's architecture consists of the following major components:
knowledge base, ideological goal trees (Carbonell 1979), rule-based natural
language generator, rhetorical devices, and a database of indexed audio/visual
elements primarily consisting of short digital movies and sound files contain-
ing music (for more architectural details than are provided in this chapter, see
(Mateas, Domike & Vanouse 1999; Mateas, Vanouse & Domike 2000)). The ar-
chitecture is depicted in Figure 2. The knowledge base contains representations
of historical events. This is the raw material out of which the ideologically-
biased histories are constructed. Examples of historical events are the First
Crusades, the invention of Bakelite, and the rise of enlightenment philoso-
phy. Ideological-goal trees represent the current ideological-bias being pursued
by the narrator. The goal-trees consist of rhetorical goals ordered by subgoal
and importance (to the ideologue) relationships. These goals are used both
to select historical events to include in the story and to “spin” the event in
an ideologically-consistent manner. The rule-based natural language genera-
tor (NLG) generates the narrative text once specific facts have been selected
and connected to make a story. The storyboard serves as a working memory
for processes that impose a narrative order on event spins created by the goal
tree. Rhetorical devices are connecting pieces of text with accompanying con-
straints on story structure. These devices are used to create narrative connec-
tions between historical events. Finally, the multimedia database contains the
audio/visual elements for the assembled documentary. Once a narrative track
has been constructed, information retrieval techniques are used to match the
“best” indexed multimedia elements to the appropriate pieces of text. Once
the multimedia elements have been selected, the resulting documentary is dis-
played, layering text-to-speech synthesis of the narrative track, and the video
and audio elements.
The audience's responses to the questions influence the machine by select-
ing and editing rhetorical goal trees, selecting a set of rhetorical devices, and
placing constraints on the storyboard. In a sense, the audience response pa-
rameterizes the machine. The responses activate structures and processes; the
machine then autonomously generates a biased history.
