The "common sense" approach as alternative to the hybrid approach


Lukas Bulwahn
 

Dear all,

I would propose an alternative approach to architecture:

(1) There is a claim on a software, for which increased confidence in
its validity is required.
(2) There are aspects that provide evidence (and hence: increase
confidence) that the stated claim holds in the software. These aspects
may be reflected in structuring the source code, combining different
means or interacting elements.
(3) An existing software or development activity can be checked to
meet a specific aspect to a certain degree.

So, the approach now is to simply address the three questions:

(Q1) What is the claim on a software for which increased confidence in
its validity is required?
(Q2) Which specific aspects provide evidence that the specific claim,
i.e., the answer to Q1, might hold?
(Q3) To which extent is a specific aspect, i.e., the answer to Q2, met
by the software or the development activity?

Of course, as the questions show, they depend on the answer of the
previous questions being answered, so they need to be addressed in
sequential order.

Does this approach make sense? Is this an alternative to the hybrid
approach that others would like to follow through with?

P.S.: I am really bad with naming; maybe there is a better name for
this approach?


Best regards,

Lukas