Computer-Aided Method Engineering: An Analysis of Existing Environments

Abstract

Analogous to Computer-Aided Software Engineering (CASE), which aims to facilitate Software Engineering through specialized tools, Computer-Aided Method Engineering (CAME) strives to support a wide range of activities carried out by method engineers. Although there is consensus on the importance of tool support in method engineering, existing CAME environments are incomplete prototypes, each covering just a few steps of the method engineering process. This paper summarizes the history and the state of the practice in CAME technology, and provides criteria-based critique on existing CAME environments, thus highlighting their strengths and weaknesses.

An Analytical Review of Process-Centered Software Engineering Environments

“…For example, scale 2 was wholly added to fit the evaluation of the coverage criterion. We have also selected and adapted certain general evaluation criteria, such as usability, from [7], which provides an analytical review of Computer Aided Method Engineering (CAME) tools. During the criteria definition and refinement process, we have strived to keep the criteria general, precise, comprehensive and balanced in order to satisfy the metacriteria of [34].…”

On the usefulness and ease of use of a model-driven Method Engineering approach

“…Regardless of the potential benefits of ME and the emergence of ComputerAided Method Engineering (CAME) technology [8], ME has never been widely used in industry [9,10]. Kuhrmann et al concluded in a recent mapping study [11] that there are hardly any reports on the practical application of ME available in the literature.…”

A Model-Driven Approach for the Design, Implementation, and Execution of Software Development Methods

“…On the other hand, in the Method Engineering literature, we can find lifecycle models of different nature [41,59,60]; nonetheless, all these models comprise generic phases that are similar to those proposed by the Software Engineering community. These phases are graphically depicted in Figure 2.3.…”

“…However, despite the great achievements of existing CAME environments (e.g., the design of the product part of methods is fully supported by a wide variety of metamodeling notations [41]), these tools still present important deficiencies: incomplete support to the Method Engineering lifecycle, weak process support, and high complexity; furthermore, these tools are affected by the negative perception of practitioners with respect to the usefulness of Method Engineering. We argue that all of these problems are the key determinants of the little use of Method Engineering in industrial contexts.…”

“…None of the existing proposals has been successful in providing a software infrastructure that allows method engineers to carry out the design, implementation, and execution of software development methods. This reality is shown in a recent study on CAME technology [41], which concludes that existing CAME (and metaCASE) environments are incomplete prototypes that only cover part of the Method Engineering lifecycle. This is one of the reasons why these tools have not achieved the expected industrial success and only MetaEdit+ [40,42] has traversed academic boundaries.…”