A deep view-point language and framework for projective modeling

Abstract

Most view-based modeling approaches are today based on a “synthetic” approach in which the views hold all the information modeled about a system and are kept consistent using explicit, inter-view correspondence rules. The alternative “projective” approach, in which the contents of views are “projected” from a single underlying model on demand, is far less widely used due to the lack of suitable conceptual frameworks and languages. In this paper we take a step towards addressing this problem by presenting the foundations of a suitable language and conceptual framework for defining and applying views for projective modeling. The framework leverages deep modeling in order to seamlessly support views that exist at, and span, multiple levels of classification. The viewpoint language was developed in the context of Orthographic Software Modeling but is more generally applicable to any projective modeling approach.

Introduction

Due to their complexity and size, enterprise architectures can only be effectively modeled using collections of intertwined views, each describing a system from distinct perspectives or viewpoints [1], [2], [3]. However, there is little consensus on what these views and perspectives should be and what kinds of properties the associated viewpoint frameworks should support. In fact, contemporary Enterprise Architecture Modeling (EAM) approaches differ quite considerably in terms of the fundamental relationships they assume between the views and the system under description, the way in which inter-view consistency is handled and how the focus of views is characterized, with significant consequences for their flexibility and/or coherence.

In a previous article [4], the authors of this paper identified seven fundamental design choices involved in the realization of EAM viewpoint frameworks based on existing literature and approaches/tools. The paper also identified the combination of choices which appear to offer the best platform for EAM in the future — namely, a rigorous framework, based on an essential underlying model which supports nestable (i.e. both abstract and concrete) views that are projective, component-centric and related by implicit and intentional correspondences. At that time no viewpoint framework possessed this combination of characteristics. In this paper we present a new viewpoint language and framework that has been specially designed to support the proposed combination of properties in a clean, efficient and uniform way. More specifically, we describe the conceptual foundations of the approach and illustrate the basic elements of a prototype environment to support it known as doreen.

Although the framework was originally conceived to support a specific view-based modeling methodology called Orthographic Software Modeling (OSM) [5], it is not restricted to this approach and can be used to support any approach based on projective views. Nor is it tied to any specific modeling language. The framework can just as easily be used with a general purpose modeling language such as the UML as with an EAM language such as Archimate [6] or RM-ODP [7]. However, some authors have pointed out that multi-level modeling languages are better suited to the EAM domain than traditional “two-level languages” such as the UML [8]. Not only do they simplify the definition and use of domain-specific modeling languages, they are able to describe instance-oriented run-time information just as easily and efficiently as type-oriented design time information within a single coherent formalism. They are therefore well suited to supporting views over an enterprise system’s full life-cycle which is an important requirement on EAM frameworks. The approach presented in this paper therefore uses, and is integrated with, a specific form of multi-level modeling known as deep modeling.

The paper is an extended version of a paper published at the EDOC 2017 conference [9]. It provides enhanced descriptions of the view type and SUM definition languages, including the underlying deep modeling environment and deep transformation language, as well as an enhanced discussion of the approach’s strengths and weaknesses. In the following section we present the background to the approach, which includes the motivation for the design choices identified above as well as an overview of the chosen deep modeling approach. The next three section then describe the three key elements of the framework: Section 3 explains how the Single Underlying Model (SUM) that serves as the source of content for the views is structured while Sections 4 View types, 5 View instances describe the languages and projection artifacts used to define and populate views at the type and instance levels respectively. Finally, Section 6 discusses the pros and cons of the approach while Section 7 concludes with some closing remarks.