Method Engineering: Towards Methods as Services
In the 90’s it was becoming apparent that a universal method that could be applied to ’any’ information system development (ISD) project is a mirage. Project specificity, differing application engineer skills and experience lead to deviations from the pre
- PDF / 89,392 Bytes
- 2 Pages / 430 x 660 pts Page_size
- 89 Downloads / 161 Views
Abstract. In the 90’s it was becoming apparent that a universal method that could be applied to ‘any’ information system development (ISD) project is a mirage. Project specificity, differing application engineer skills and experience lead to deviations from the prescriptions made by given methods. This was revealed by several survey based studies (e.g. [6], [1], [7], [4]). For example, a survey of method use in over 100 organizations’ [4] shows that more than 2/3 of the companies have developed or adapted their methods in-house. Also 89% of respondents believed that methods should be adapted on a project to project basis. These observations raised the need for techniques to rapidly build methods adapted to engineer preferences and project situations. The area of Method Engineering (ME) attempts to provide these techniques. Method engineering represents the effort to improve the usefulness of systems development methods by creating an adaptation framework whereby methods are created to match specific organisational situations. There are at least two objectives that can be associated to this adaptation. The first objective is the production of contingency methods, that is, situation-specific methods for certain types of organisational settings. This objective represents method engineering as the creation of a multiple choice setting [3]. The second objective is one in which method engineering is used to produce method “on-the-fly”. Situational method engineering [5] is the construction of methods which are tuned to specific situations of development projects. Each system development starts then, with a method definition phase where the development method is constructed on the spot. Rapid method construction is predicated on the reuse of existing methods. Thus, methods are modularised into components that are stored in a repository. ME used the notion of a meta-model for developing method components that could be purely product or process components, or integrated product and process components. These components are kept in a method repository from where they can be retrieved. The retrieved components can be adapted and ‘put together’ to form a coherent whole method through an assembly process. This composition aims to build methods that address project concerns and fit with project specificity. From the aforementioned, it can be understood that the three key issues of method engineering are, (a) defining method components, (b) finding and retrieving components and (c) assembling components. The talk will survey the results achieved so far along these three issues. It can be seen that ME provides engineering capability but, we believe, that it needs to be augmented with good usage properties. Specifically it is limited as follows: Q. Wang, D. Pfahl, and D.M. Raffo (Eds.): ICSP 2008, LNCS 5007, pp. 10 – 11, 2008. © Springer-Verlag Berlin Heidelberg 2008
Method Engineering: Towards Methods as Services
11
Despite a common acceptance of a modular vision of methods, every method engineering approach has developed its own
Data Loading...
