1.2 Productivity Changes After Initial Introduction
Although when a DSM solution is introduced the productivity difference is most
visible compared to earlier approaches, possible productivity gains are not limited to
its introduction.
The Important Learning Aspect Usually productivity studies comparing
traditional and newapproaches don??™t recognize the learning aspect: once modelers get
acquainted with the new approach as well as with the earlier practices, the difference
can be expected to become even bigger. This is especially important for analyzing
DSM cases, as the use of modeling for complete code generation is new for many
developers. For example, in the DSM case discussed in Chapter 6 dealing with the
insurance domain, the company estimated its productivity increase as 300% compared
to earlier manual programming practices. This estimation was done after
specifying the ?¬?rst 30 insurance products out of more than 200. Since the software
modeling was new to almost all of the developers, it was expected that productivity
would continue to increase later too. Unfortunately the company did not collect data
from these later phases.
Productivity in Maintenance Phases Often most development effort and time
go into the maintenance phase. There are roughly three main kinds of maintenance
work: bug corrections, responses to new or changed requirements, and responses to
platform or environment changes.
Pages:
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67