Michael Gort <mail18081@pop.net> says:
>In fact, we have experienced a phenomena that Ken Cooper of Pugh-Roberts
>calls the lost year. The project is reported to be 90% completed, then
>seems to hang at 90% for as much as a year or more. Dynamics models
>usually show that the "lost year" is tied to continuing to a) discover
>work that was not specified or considered in sizing, and b) to ongoing
>discovery of rework (which can equal as much as the full project
>estimation!). If something is not done about the quality issue, fewer and
>fewer developers are able to work on "new", and instead must constantly
>focus on fixing bugs, adding enhancements (frequently, features that were
>dropped to get the product out the door). A separate feedback loop
>results from the developers inability to turn products over to a separate
>support group. The interrupt-driven nature of support is antithetical to
>the social stream of consciousness nature of software team development.
This 'lost year' is a function of two elements ot the development cycle:
1. Integration management (or the determination of the systems resources,
including human intelligence, behavioural modality traits and the like) and
2. lack of earned value management in the project management methodology.
The latter is an important learning issue, especially in software
projects. It is usual for some version control and project management
software to be used to control the project. It is, however, unusual for
the team to understand the implications of the methodologies (in
particular the Project Managers' Body of Knowledge) which are built into
it.
-- John Zavacki jzavacki@wolff.com Wolff Group, Inc. 800-282-1218 http://www.wolff.comLearning-org -- An Internet Dialog on Learning Organizations For info: <rkarash@karash.com> -or- <http://world.std.com/~lo/>