This was
done by comparing the number of failed tests with manual approaches and with
domain-speci?¬?c languages and generators. The study found signi?¬?cantly fewer errors
in the domain-speci?¬?c approach. Although the generator was newly created for the
task, the acceptance tests found 50% fewer errors from systems built using the
domain-speci?¬?c approach than from systems built manually. The differences in the
average performance of the subjects were statistically measured and found signi?¬?cant
at a con?¬?dence level of 97%.
2.2.2 Mapping to Requirements and Problem Domain
By providing support for problem domain concepts, DSM extends the role of
languages from traditional design and implementation use towards requirements
capture. In many cases, the domain-speci?¬?c models express things that in traditional
approaches would be close to the requirements a customer speci?¬?ed. Consider
here again the mobile application example from Chapter 1. The requirements on
application functionality, type of widgets to be used, and user navigation are directly
expressed in the domain-speci?¬?c model (Fig.1.6). This close mapping from models to
requirements enables customers and other domain experts to participate in
development work.
Participation with Customers and Domain Experts Speci?¬?cations made
with domain terms are usually easier to read, understand, remember, validate, and
communicate with.
Pages:
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73