Information Technology Reference
In-Depth Information
AI or “classical” AI—its principles sit beside both (Th orisson et al., 2004). In fact,
because CDM is intended to address the integration problem of very broad cogni-
tive systems, it must be able to encompass all variants and approaches to date. It is
unlikely that a seasoned software engineer will find any of the principles presented
objectionable, or even completely novel for that matter. But these principles are
custom-tailored to guide the construction of large cognitive systems that could be
used, extended, and improved by many others over time.
2.3.2.4.1
Advantages
Modularity at its center, where functionalities of the system are broken into
individual software modules.
CDM's principle strength is in simplifying the modeling of complex, multi-
functional systems requiring architectural experimentation and exploration of
subsystem boundaries, undefined variables, and tangled data flow and control
hierarchies.
2.3.2.4.2
Disadvantages
Not proliferated into other industry or areas other than AI.
2.3.2.4.3
Suitability
CDM is a methodology for designing and implementing interactive intelli-
gences, and it is mostly suitable for building large cognitive robotics sys-
tems, communicative humanoids, facial animation, interdisciplinary collabo-
ration support, coordination of teams, and large-scale systems integration in AI.
It is most applicable for systems with ill-defined boundaries between subsys-
tems, and where the number of variables to be considered is relatively large. In
the current state of science, primary examples include ecosystems, biological
systems, social systems, and intelligence.
2.4
SOFTWARE DEVELOPMENT PROCESSES CLASSIFICATION
The classification of traditional software development processes can be done in
many different ways; however, here the models discussed in Section 2.2 are viewed
from complexity and size of a project. Table 2.1 shows the classification based on the
suitability of size and complexity of project. The gray areas shown in Table 2.1 depict
the nonsuitability of the given software process depending on the size and complexity
of the project. This does not mean the process cannot be used, but knowing the nature
of the process or model's best results may not be obtained.
Search WWH ::




Custom Search