Less Speci?¬?cation Work Needed. Keeping speci?¬?cations at a signi?¬?cantly
higher level of abstraction than traditional source code or class diagrams means less
speci?¬?cation work. As the language need ?¬?t only a speci?¬?c domain, usually in only
one company, the DSM can be very lightweight. It does not include techniques or
constructs that add unnecessary work for developers. Consider here, for example, the
difference between the amounts of modeling work needed in the UML and DSM
approaches presented in the mobile application case in Chapter 1.
Less Testing Needed and Many Typical Errors Disappear. In DSM, a large
portion of the testing is effectively done before the modeling stage as the language
contains the rules of the domain. The rules need to be in the modeling language to avoid
generating code from models that are full of errors. Similarly, most typical errors in
manually written programs, such as typos, missing references, using variables not yet
initialized, and errors in memory allocations, no longer occur because the code is
generated. Application developers thus don??™t need to test for these anymore as they are
effectively tested by the language.
No Need to Change the Generated Code. The generator is made by your own
expert developer, not a vendor, so it already produces the code you need.
Pages:
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119