Modeling Forms of Information Derivation in Modified WebML+

Rachatrin Tongrungrojana [HREF1], Faculty of Engineering [HREF2], University of Technology, Sydney [HREF3]. rachatrin.tongrungrojana@uts.edu.au

David Lowe [HREF4], Faculty of Engineering [HREF2], University of Technology, Sydney [HREF3]. david.lowe@uts.edu.au

Abstract

The ability to reliably and consistently develop systems that utilise internet and Web technologies has become increasingly important. These systems are typically both functionally complex and information-rich, and have a number of unique characteristics that should imply specific changes to the development processes, methods and models that are adopted. One aspect of the development of Web-enabled systems that has received increasing attention is information modeling, particularly with respect to aspects such as navigation and content models. These models have, however, typically focused on modeling at a relatively low-level and have failed to address higher-level aspects, such as architectural and even business process modeling. We propose WebML+, a set of formal extensions to an existing modeling language (WebML) which addresses these concerns.

In this paper we present propose a modification to WebML+ that facilitates improved modeling (and hence understanding) of the relationships between information units within the organisation. Specifically, WebML+ includes the concept of information being derived from other information units. We look at enhancing the modeling of these relationships, to capture the different forms of derivation that can occur.

Introduction

A key element in the development of effective Web systems is the availability of an appropriate Web modelling language. Most existing languages focus on understanding the structure of the information space (see, for example, WebML (Ceri et al., 2000)). These approaches however have rarely addressed the connection between these detailed design aspects and the broader information environment, particularly in terms of the flow of information between the system, the organisation, and external entities. It is this higher level modelling that is the focus of our work, and which has resulted in the development of a notation that bridges the gap between business models and information design. The details of this modelling language (which extends WebML, and hence we have labelled WebML+) are provided elsewhere (Lowe and Tongrungrojana, 2003, Tongrungrojana and Lowe, 2003a).

In this paper we propose a modification to WebML+ that facilitates improved modelling (and hence understanding) of the relationships between information units within the organization.  Specifically, WebML+ includes the concept of information being derived from other information units.  We look at enhancing the modelling of these relationships to capture the different forms of derivation that can occur. We begin with a brief overview of WebML+ and then describe the modifications that we are proposing, as well as providing illustrative examples.

WebML+

WebML+ enables developers to express the core information features of a system at a high level of abstraction, without committing to detailed architectural designs. It can be considered as an extension to WebML (see (Ceri et al., 2000) and www.webml.org). The purpose of WebML+ modeling is to define both the internal and the external information flows within an organisation.

As with WebML, we have defined both a graphical notation and an XML-based formal notation for representing WebML+ models (though we do not show the formal XML DTD here). The graphical notation is designed to allow it to be effectively communicated to non-technical members of development teams. WebML+ supports a numbers of different modeling entities. The full formal definition of these entities and their inter-relationships is given in (Lowe and Tongrungrojana, 2003).

Modeling Entity	Example
Actor units are defined to show roles that users play with respect to the system and show how users participate with information flows.
Supplied information units present information about a single information object that is provided by actor.	Transient supplied info	Persistent supplied info
Derived information units present information units that are derived from other information units (i.e. the units are not supplied by actors). This derivation may be carried out as a process for generating the underlying content.
Information flows describe that flow of information within the system and between the system and the external actors.

WebML+ support for extended source of derivation

By introducing the concept of derivation, WebML+ supports understanding the relationships between different information units. This derivation may ultimately be implemented as a process that either dynamically or statically generates the underlying content. We have extended this concept by identifying different forms of derivation that can occur.

Direct derivation describes a process of an information unit being derived directly from other information units. In other words, the derived information is dependent directly on the source information. This is represented as follows:

Behavioural derivationdescribes the situation where an information unit is derived not from an information unit itself, but rather from the way in which that information is made available or accessed.  For example, by analyzing the timing of a user query we may be able to derive information about a user's behaviour. In this case the derived information is likely to not have been the main focus of the actor providing the source information.

Meta derivationoccurs when an information unit is derived from meta-data associated with another information unit, rather than the unit itself. For example, type of Web browser, geological location, etc.

Example

Figure 1 shows an example of a WebML+ model: Library Online is Web version of the library catalogue that allows users to enquire for information about books, papers, etc. through Web interface. In this example, we have 2 actors: librarians as internal actors and users as external actors. Internal actors such as librarians manually provide information (e.g. book and publication information) to the system. With the external actors, users provide user information (such as personal identity and contact information), as well as information query to the system while they receive search result (which is, in turn, derived from information query itself, user information, and information from book database), books suggested by the system (which are derived from information from book database and user information and profile), and user policy.

So, let us consider what is represented in the organization. The organization boundary (shown as a dashed geometrical polygon) encloses a set of information units which represent coherent and cohesive domains of information that are managed or utilized by the organization while the system boundary (shown as a dotted geometrical polygon) encloses only the set of information units that are utilized by the system under consideration. Some information units are provided directly by actors (e.g. query and user info from users and book info from librarians) while many information units are derived from other units rather than being provided explicitly (e.g. Search result is derived from information query, user information, and information from book database).

In this example, in particular, we focus on our proposed modification. Several direct derivations occur in the example, such as CreateSearchResult and Create SuggestedBook derivation. UserBehaviour (which is behavioural derivation) is derived from the analysis of timing of a user query. UserLocation is derived from meta-data of UserInfo (which is geological location in this case so we can define this derivation is meta derivation).

Figure 1 Typical WebML+ Model

Conclusion and related work

In this paper, we have presented a modification to WebML+ that supports a richer model of the inter-relationships between information units. As we have discussed, the WebML+ approach is built around the notion of information flows with in the organisation at the level of understanding business processes. We argue that this enhancement facilitates improved modelling and hence understanding of the relationships between information units within the organization.

Despite a major aspect of WebML+ is the ability to define the information flows within an organisation, system and its context, this model also is able to form a bridge between the business modelling and the lower level information modelling. This issue is well explained in (Tongrungrojana and Lowe, 2003b).

References

Ceri, S., Fraternali, P. and Bongio, A. (2000). "Web Modeling Language (WebML): a Web modeling language for designing Web sites" in Proceedings of WWW9 Conference, Amsterdam.

Lowe, D., Tongrungrojana, R. (2003). "WebML+: a Web modeling language for modeling architectural-level information flows" in Proceedings of The Twelfth International World Wide Web Conference, Budapest.

Tongrungrojana, R., Lowe, D. (2003). "WebML+: a Web modeling language for modeling architectural-level information flows" in International Journal of Web Engineering and Technologies.

Tongrungrojana, R., Lowe, D. (2003). "WebML+: a Web modeling language for forming a bridge between business modeling and information modeling" in Proceedings of The Fifteenth Information Conference on Software Engineering and Knowledge Engineering, San Francisco.

Hypertext References

HREF1

http://www.rachatrin.com/

HREF2

http://www.eng.uts.edu.au/

HREF3

http://www.uts.edu.au/

HREF4

http://www.eng.uts.edu.au/~dbl/

Copyright