An architectural perspective plays the same role as an architectural viewpoint, except that: (1) it may apply to more than one view, whereas a viewpoint applies to exactly one, and (2), a perspective addresses exactly one quality-related concern (or quality property), whereas a viewpoint may address one or more concerns. The other differences suggested by WER have been shown in the previous section to be non-issues. Assuming that the notion of architectural perspective is useful, can it be used in combination with IEEE 1471, given these differences? There are two options: (a) apply perspectives as a part of an architecture method, outside the scope of IEEE 1471, but leading to IEEE 1471 conforming ADs; or (b) assimilate the notion of perspective into the IEEE 1471 conceptual framework, adjusting the IEEE 1471 requirements or the definition of perspective, or both, to accommodate.
Under (a), (1) and (2) do not directly arises as issues. However, under (a) the problem with perspectives is that, as defined, their use may compromise the ability of a resulting AD to conform to IEEE 1471. Consider the requirement:
Each stakeholder and each concern identified in an AD in accordance with 5.2 shall be addressed by at least one viewpoint selected in accordance with 5.3. [IEEE 1471, clause 5.3].
Replaced/Superseded by document(s)
|File||MIME type||Size (KB)||Language||Download|
This note analyzesWoods’ and Rozanski’s notion of architectural perspective in light of the conceptual
framework of IEEE Std 1471.