First,
modeling languages can include correctness rules of the domain, making it dif?¬?cult,
and often impossible, to create illegal or unwanted speci?¬?cations. Elimination of bugs
in the beginning is far better than ?¬?nding and correcting them later. Finding
corrections is usually easier and cheaper when done earlier. Models can also be
checked and analyzed to detect errors or ?¬?nd speci?¬?cations that lead to poor
performance in the product build. Second, generators along with framework code
provide a mapping to a lower abstraction level, normally code, and the generated
result does not need to be edited afterwards. By automating the mapping from a
problem domain to a solution domain we also reduce the risk of the implementation
not corresponding to the solution speci?¬?ed in the problem domain.
As with the productivity gains, the improvements in quality do not come
automatically. Someone must de?¬?ne the modeling languages, generators, and
supporting frameworks. In DSM, a key reason for the better quality is that a large part
of all applications is effectively built by the more experienced developers??”those who
created the DSM solution. We next describe in more detail how DSM contributes to
improving quality.
2.2.1 Measuring Quality Improvements
Quality can be measured in many ways, such as the number of errors found, how
reliable the product is, how much maintenance effort is needed, or how fast the
application performs the required operations.
Pages:
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71