A key challenge to move forward the state of the art in games-based learning systems is to facilitate instructional content creation by the domain experts. Several decades of research on computer aided instruction have demonstrated that the expert has to be deeply involved in the content creation process, and that is why so much effort has been devoted to building authoring tools of all kinds. However, using videogame technology to support computer aided instruction poses some new challenges on expertfriendly authoring tools, related to technical and cost issues. In this chapter the authors present the state of the art in content creation for games-based learning systems, identifying the main challenges to make this technology cost-effective from the content creation point of view.
Content Creation In Videogames
Video game content may be divided roughly in two different groups: assets and behaviours. The term assets usually denotes those elements that go into a game, such as the artwork (textures and 3D models), sound effects, music, and, generally speaking, every object that is presented to the user. On the other hand, the way in which the objects behave constitutes the second kind of content and is sometimes referred to as dynamic content. Both kinds of content are usually handmade, therefore their cost constitutes an important part of the game budget. To compound matters, the situation is getting harder, because of the ever growing hardware capabilities that allow video games to present more and more objects with higher resolutions, and therefore requires more and more people involved in the creation of all this content.
The creation of assets involves artists using tools such as 3D StudioTM, MayaTM and PhotoshopTM to generate 3D models and textures. Depending on the game, the amount of this content varies from just a few models to several Gigabytes of stored files (Gillen 2005). In order to alleviate the cost, some effort has been made to create algorithms that build part of them procedurally. Procedural content generation was used by early games as the only way to fit vast amounts of data onto the small mediums available at that time. Nowadays the goal is not just to save disk space but also man power and therefore budget. Examples of assets procedurally generated are textures, terrains or trees; recent developments like Spore are also exploring dynamically generated animations, and Diablo III, Hellgate: London or Borderlands have random level generation.
On the other side, behaviours constitute the dynamic part of the game, because they define what players can do with the objects in the environment and how they react to their actions. These behaviours are usually created by AI and script programmers.