WSDM - Web Semantic Design Method
WSDM is an audience driven design method for Web Applications. It consists of a five-phase design, taking as a starting an explicit elicitation of the users and ending with the actual implementation (generation). More than other methods, WSDM is a methodology, i.e. it not only provides modeling primitives that allow a web developer to construct models that describe the web site/application from different perspectives and at different levels of abstraction, but it also provides a systematic way to develop the web application. In the figure below, you can see the five design phases, and the output model(s) produced in each phase.
Mission Statement Specification
In the first phase of WSDM, the mission statement for the web site is expressed. In the mission statement, the purpose, the subject and the targeted users of the web site are specified. The mission statement establishes the borders of the design process. In the rest of the method it will allow to decide what information or functionality to include or not, how to structure the information and how to present it. In addition, the mission statement is useful for validation, i.e. to check if the web site has achieved the formulated purpose. The mission statement is formulated in natural language.
The mission statement formulated in the first phase is only a first and very incomplete description of the system that should be developed. Because WSDM is an audience-driven design method, the first concern that is elaborated is the set of target users. WSDM takes into account the fact that different types of visitors may have different needs and requirements, and thus may require different support (e.g., in a university website, 'students' and 'researchers' clearly have different goals and purpose to visit the website). Therefore, the target users identified in the mission statement are refined into so-called audience classes. This is done by means of two sub phases: the Audience Classification and the Audience Characterization. During Audience Classification, the different types of users are identified into more detail and classified: users with the same (information and functional) requirements are grouped together into so-called audience classes. When the set of requirements of a particular audience class is a subset of the set of requirements of another audience class, we call this first audience class an audience subclass of the latter. An audience subclass thus has the same, and some additional requirements as its audience superclass. Based on the subclass relationships, the audience classes can be classified into a so-called audience hierarchy, which will later on be used to derive the basic navigational structure (see Conceptual Design). For each audience class, navigation and usability requirements are stated as well. During Audience Characterization relevant characteristics are specified for each audience class. Next to the fact that different types of users may have different information and functional requirements, it may be necessary to represent the (same) information or functionality in different ways to different kinds of users. This depends on the characteristics of the users.
During the previous phases, mission statement specification and audience modeling, the information-, functional-, usability- and navigational requirements as well as the characteristics of the potential visitors were identified and different audience classes are defined. The goal of the Conceptual Design is to turn these informal requirements into high level, formal descriptions which can be used later on to generate (automatically or semi-automatically) the web system. During Conceptual Design, we concentrate on the conceptual “what and how” rather than on the visual “what and how”. The conceptual “what” is covered by the Task & Information Modeling sub phase and deals with modeling of the content and functionality of the web system; the conceptual “how” is covered by the Navigational Design sub phase and specifies the conceptual structure of the web system and the navigation.
Taks & Information Modeling
The purpose of the Task & Information Modeling is to model in detail the different tasks the members of each audience class need to be able to perform, and to formally describe the data and functionality that is needed for those tasks. The tasks that a member of an audience class needs to be able to perform is based on the requirements formulated for the audience class during Audience Classification, i.e. for each information and functional requirement formulated for an audience class, a task is defined that allows to satisfy this requirement. Each task is modeled into more details using an adapted version of the task modeling technique Concurrent Task Trees (CTT); originally developed in the context of HCI. Essentially, in CTT, tasks are decomposed into subtasks until elementary tasks are obtained. In addition, temporal relationships between subtasks allow to indicate the order in which the subtasks need to be performed (e.g. sequential, parallel, etc). Once a task decomposition is completed, each elementary task is modeled by a so-called object chunk. An object chunk formally describes the information (and/or functionality) needed to complete this task. Initially, Object Role Modeling (ORM) was used for this purpose, but to better suit the needs of the semantic web, WSDM now uses OWL (with some specific modeling concepts to support functionality). The use of OWL allows the use of (one or more) existing ontologies, and facilitates the (automatic) generation of semantic websites (see Semantic Annotations), effectively enabling the Semantic Web.
The goal of the Navigational Design is to define the conceptual structure of the web system and to model how the members of the different audience classes can navigate through the web system and perform their tasks. Because of the audience driven approach of WSDM, a navigation track is created for each audience class. Such an audience track can be considered as a sub-site containing all and only the information and functionality needed by the members of the associated audience class. The internal structure of an audience track is derived from the task models made for this audience class. A navigational model is created by means of components (also called 'nodes') and links: components are conceptual navigation entities, referring to information/functionality (by means of the object chunks) that logically belongs together; link are connections between components. WSDM distinguishes four link types: structural links (denoting pure structure), semantic links (based on semantic relationships), navigational aid links ('shortcut') and process-logic links (part of a workflow). These type of links can for example be used when generating the presentation the links: depending on their purpose, links may have a different presentation.