Designing and maintaining Web applications is one of the major challenges for the software industry of the year 2000. In this paper we present Web Modeling Language (WebML), a notation for specifying complex Web sites at the conceptual level. WebML enables the high-level description of a Web site under distinct orthogonal dimensions: its data content (structural model), the pages that compose it (composition model), the topology of links between pages (navigation model), the layout and graphic requirements for page rendering (presentation model), and the customization features for one-to-one content delivery (personalization model). All the concepts of WebML are associated with a graphic notation and a textual XML syntax. WebML specifications are independent of both the client-side language used for delivering the application to users, and of the server-side platform used to bind data to pages, but they can be effectively used to produce a site implementation in a specific technologica...

Many Web sites include significant and substantial pieces of information, in a way that is often difficult to share, correlate and maintain. In many cases the management of a Web site can greatly benefit from the adoption of methods and techniques borrowed from the database field. This paper introduces a methodology for designing and maintaining large Web sites based on the assumption that data to be published in the site are managed using a DBMS. We see the process of designing the site as the result of two intertwined activities: the database design and the hypevtex't design. Each of these is further divided in a conceptual design phase and a logical design phase, based on specific data models. A new logical data model, called ADM, is used to describe the structure of a Web hypertext. It is page-oriented, in the sense that the main construct is the one of page-scheme, providing an intensional description of a class of pages in the site. Based on the ADM scheme of the site, we introduce a language, called PENELOPE, that allows to automatically generate HTML pages starting from the database content. PENELOPE is also able to correlate different pages in a complex hypertext using a suitable URL invention mechanism to guarantee reference integrity. ADM and PENELOPE strongly support site maintenance: the first provides a concise description of the site structure; it allows to reason about the overall organization of pages in the site, in order to evaluate the effectiveness and efficiency of the chosen structure, and possibly to restructure it; at the same time, PENELOPE alleviates the burden of managing HTML files by hand, and guarantees link consistency in presence of updates and reorganizations.

Many of the Web applications around us are data-intensive; this term indicates applications whose main purpose is presenting large amount of data to their users. Most of the sites for on-line trading or e-commerce are data-intensive, as well as most “institutional ” sites (describing private and public organizations) or digital libraries. Several commercial Web development systems enable the rapid development of data-intensive

Starting with a conceptual model when designing a web site is the state of the art. A conceptual model helps to grasp and structure the problem domain and is the first step towards a formal representation of the web site, provided that the chosen technology has a formal foundation. Applying the extended entity relationship driven EER/GRAL-approach to specifying graph classes, we show that graph technology can be utilised to ensure a coherent and consistent usage of a conceptual model and its instances for defining and generating an arbitrary complex web site. During this process, graphs are used as repository structures in conformance with the conceptual model, allowing for descriptive graph queries to define the contents of the web pages. Along withthe application of XSL (extensible style sheet language) as a means to foster separation of content and layout, this approach ensures a permanently consistent website. Some examples are given.

Web sites are progressively evolving from browsable, read-only information repositories to web-based distributed applications. Compared to traditional web sites, these web applications do not only support navigation and browsing, but also operations that have affects their contents and navigation states. Compared to traditional applications, web applications integrate operations with the built-in browsing capabilities of hypermedia. These novelties make web application design a complex task that requires the integration of methods and techniques developed in different "worlds". This integration is achieved in this paper by extending and customizing the Unified Modeling Language (UML) with web design concepts borrowed from the Hypermedia Design Model (HDM). Hypermedia elements are described through appropriate UML stereotypes. UML diagrams are also tailored to model operations and relate them with hypermedia elements. The approach is exemplified by describing the design of a web-based conference manager.

this article we describe how developers can incorporate hypertext functionality into computational applications. What benefit do users gain from hypertext support in computational applications? Managing the myriad of interrelationships in a computational application's knowledge (data and calculated information) is difficult for a user. It can be improved by streamlining access and increasing user comprehension through interface enhancements (e.g., visualization). Augmenting an application with hypertext support results in new ways to view and manage the application's knowledge, by navigating among items of interest and annotating with comments and relationships (links). Unfortunately, many computational system developers view hypertext only in terms of accessing and managing documents (or smaller units of static information) . Such display-oriented behavior characterizes the majority of hypertext systems. SEPIA (see Thring et al. and Streitz in this issue), KMS [1], Aquanet and VIKI

The Dexter Hypertext Reference model is well suited to modelling anchor-based hypermedia systems and static hypermedia structures. But it is less clear that Dexter is adequate for systems whose linking is based on embedded references like the World Wide Web (WWW), nor for modelling the dynamic aspects of contemporary hypermedia systems like DHM and Microcosm. This paper proposes a new Dexter-based extensible objectoriented model designed to cover a broader spectrum of the features of contemporary hypermedia systems. The model introduces two new concepts, LocationSpecifiers and ReferenceSpecifiers, which let us model links as references embedded in documents as well as links as objects in separate databases. This suggests the idea of new systems that could support both styles as one step toward integrating global networked information sources with application-bridging systems on local hosts. In addition, our model is better equipped to handle dynamic hypermedia structures. As an example...

E-commerce and m-commerce demand for full-fledged, increasingly complex applications which need to offer ubiquitous access in terms of the anytime/anywhere/anymedia paradigma. From a software engineering point of view, the development of such ubiquitous web applications requires proper modeling methods in order to ensure architectural soundness and maintainability. Recently, web modeling methods started recognizing this fact by providing first concepts for dealing with ubiquity. In this paper, two of these modeling methods are compared, identifying their strengths and shortcomings. As a prerequisite, an evaluation framework is introduced, using the notion of customization as the uniform mechanism to enable ubiquity. Customization adapts a web application towards a particular context which reflects the environment the application is running in. To enable a holistic view on the development process of a ubiquitous web application, customization is regarded as a new modeling dimension, influencing all other tasks of ubiquitous web application development.

Abstract This article presents OO-HMethod, an extension of the OO-Method conceptual modelling approach to address the particulars associated with the design of web interfaces. It is based on the OO-Method class diagram, which captures the statics of the system. The design of the interface appearance and the navigation paths are driven by the user navigation requirements. To achieve its goal, OO-HMethod adds several navigation and interface constructs to the OO-Method conceptual model, which define the semantics suitable for capturing the specific functionality of web application interfaces. A new kind of diagram, the ’Navigation Access Diagram ’ (NAD) is introduced. All the concepts represented in the NAD are stored in a repository, and from there a functional interface is generated in an automated way. One of the main contributions of this paper is not the proposal of yet another method for web modelling but the extension of an existing conceptual modelling approach. 1

The web is more and more used as a platform for fullfledged, increasingly complex information systems, where a huge amount of change-intensive data is managed by underlying database systems. From a software engineering point of view, the development of such so called DataWeb applications requires proper modeling methods in order to ensure architectural soundness and maintainability. The goal of this paper is twofold. First, a framework of requirements, covering the design space of DataWeb modeling methods in terms of three orthogonal dimensions is suggested. Second, on the basis of this framework, eight representative modeling methods for DataWeb applications are surveyed and general shortcomings are identified pointing the way to nextgeneration modeling methods.