The pilot project
should be a real application that is important to the business, in the sense that it would
have to be built anyway and will form part of a real product. It should not, however, be
on the critical path or be strongly coupled with other projects. Wherever possible, the
pilot project should be one that is already well understood: the kind of job that you
might give to a new developer.
The key criteria for the pilot team are motivation and discipline. While the team
need not be gung ho DSM enthusiasts, they must be willing to give it a go, and able to
cope with the setbacks and delays that accompany any new venture. Since the
modeling language and generators will be un?¬?nished and changing rapidly, the pilot
team must be disciplined in following instructions, updating to newer versions of the
language and generators, and reporting the issues they encounter.
13.2.1 Management Support
The ?¬?rst steps with DSMcan often be performed within the bounds of existing resource
allocations, but full implementation will normally require management support. Since
there is usually little chance for a teamto select or instruct its ownmanagement, wewill
restrict ourselves to looking at what can be requested from and expected of the existing
management. In the broadest terms, we are looking for support from the management
for the process of evaluating and implementing DSM, for as long asDSMappears to be
a valid solution for this area.
Pages:
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625