The development of multimedia documents is a complex and expensive process. Since documents make use of different types of media such as images, text, video and audio they can not automatically adapt to the specific environment in which they are displayed on by the user such as a smart phone or laptop.
Joost Geurts, from Centrum Wiskunde & Informatic (CWI) in Amsterdam, has done research on a model that automatically adapts multimedia documents to their environment. On February 3, he will receive his PhD at TU Eindhoven for his dissertation "A Framework Document Engineering and Processing Model for Multimedia Documents." With the model he describes in his thesis multimedia documents can be developed much more cost effectively and become more widely accessible to a broader audience.
"For example, text-based documents can be automatically adjusted to the size of a page because a line of text can be automatically hyphenated and continue on the next line or page if necessary. A similar automated method unfortunately does not exist." explains Geurts. According to Geurts it is therefore highly complex to develop multimedia documents that can be rendered suitable for different devices and users. For each environment, the document must be re-adjusted. "The location of a photo and text and where it should be positioned or not, and whether they might also be connected to each other in time, can really only be made by someone who understands the semantic relationship between text and photo. Furthermore, it must also be attractive and clear. As a result, professional design knowledge is needed and that is expensive."
In order to simplify the use and application of multimedia documents Geurts has formalized professional design knowledge and incorporated it in a system that automatically adapts the document to environment in which it is presented. The underlying model was based on a common model used for text which explicitly separates content from formatting. Similar to technologies available for text, Geurts uses the model of a "style sheet" in order for the explicit design knowledge to be reused. The advantage is that multiple documents can use the same stylesheet and this saves production costs. The software implementation of the model automatically generates documents in different versions including SMIL (Synchronized Multimedia Integration Language) and SVG (Scalable Vector Graphic). SMIL is a markup language for multimedia presentations that was developed at CWI and is based on XML. SVG is written in XML and used in conjunction with CSS.
At present, multimedia documents are hardly used in daily life except for professional markets such as the advertising industry and mobile phone providers that offer services to their users to sync pictures and text. Geurts stresses that multimedia documents are highly effective in conveying messages, but that current technologies do not provide opportunities for broader use. With his research, he hopes to bring about change in this area.