Development of the informative classification scheme for information system artifacts would be highly useful for design researchers in focusing and organizing their research projects and identifying gaps. There have been few dated attempts at IS classification mostly focusing on intra-organizational systems. This chapter stresses the need for newer frameworks, which would accommodate for recent developments in IS from the design-type research perspective. The chapter outlines one possible approach, which incorporates individuals, groups, organizations and markets as possible components. Classification could span through the layers of the representational framework presented earlier to produce the families of meta-requirements and synthetic and technological meta-systems. Design research frontier helps in identifying possible developments from the existing meta-systems towards true future system forms. Along this path design researchers are expected to encounter phantom forms.
TopDesign-type research in IS aims at discovering generic forms for such systems or their constituent components, which would be beneficial for improving performance of individuals, groups, organizations, and markets. These forms, as pointed earlier, are essentially abstract entities, which could be instantiated in particular environmental contexts. While design-type science is, essentially, a C-type of science, design activities begin by producing the proper description of the (generic) problem context and examination of the existing solution forms. In order to provide guidance for researchers and help them organize their projects, a descriptive typology of the existing problem contexts and solutions would be highly valuable. In other words, a general typology of existing information system solutions, including those widely used in practice, as well as the ones proposed by researchers and not-yet adopted by the practitioners, would be very much desirable.
An ideal model for classification of the IS would have a form of a hierarchy. Hierarchical typologies are well-organized and have a relatively low level of complexity. It would have been easier for researchers to position their work somewhere along the branches of such a hierarchy, perhaps by adding an additional “leaf” to the terminal branches. This would have been very much akin to the way different living forms are classified in the biological domain. Each type of organism is classified as particular species. At a higher level the similar forms are aggregated into genera, and so on. Thus, as one progresses towards higher levels in the hierarchy, the forms become more and more abstract. Applied to the IS domain, this would translate into the hierarchies of system forms of varying levels of abstractness. For example, one could position an active decision support system form as a subclass (species) of the DSS genus, which, in turn would be a subtype of a superclass of Information Systems, which, in turn is a subclass of computing artifacts. Then, just like in the biological domain, the system types could be informatively described in terms of their features, structure, and behavior. The design researcher, could then be viewed as a discoverer of new “species” of IS, rather than an inventor of an artifact.
Unfortunately, an adequate way to arrive at such a clean classification hierarchy seems to be much more difficult compared to the biological model. The way biological evolution is working is not quite similar to the technological progress. Every species has one and only one parent genus. Interchange of form and behavior aspects between the individuals of different species is impossible, as these do not interbreed or their offspring is fertile. The biological taxonomy is well-aligned with the chronological order of the succession of living forms. The difficulty with developing a hierarchical classification schema for IS lies, partially in the fact that classes of systems may actually be regarded as subclasses of several more generic types. There is no particular rule in technological evolution that prohibits the mixing of various higher abstraction level form elements to produce a more specific subform. For example, a knowledge-based decision support system form may be regarded as an offspring of general DSS and general expert system forms, and thus, it would be inheriting some features and behaviors from both parents. This type of permitted relationships between classes is known as multiple inheritance, and it could lead to quite complex structures. In this regard, in object-oriented development the designers and programmers are well-advised to refrain from using multiple inheritance; in fact it is forbidden in later languages, including Java and C#, as opposed to earlier ones (e.g. C++). Chronological approach would not work either, as the evolution of technology is not quite as orderly as the natural evolution. It would be highly doubtful to have the applications that had historically appeared first at the root of the classification hierarchy. Indeed, it would be inadequate to put early payroll applications as the most generic type of information systems.