|
Data Management has a very sensible and straightforward approach to this process: we read the contract, we read the associated sections of the contract and the scope of work or the scope of objectives. This careful read of the contract provides clear guidance on what is to be developed, and what capabilities must support the developmental item. It is from the requirements in the contract that hardware and software items are described, and the extrapolation from which that information is made. Once the CI's have been identified, the processes of creating the software and hardware hierarchy, developing the end items, creating and managing the HW/SW baseline, releasing the items through the version control process, and authoring the necessary hardware and software specs begins. An integrated hierarchy of software and hardware items and their associations and relationships provides a clear departure point for CI development. CM and DM are arts as well as scientific disciplines. The art of software CI identification process is to understand the operation of the proposed system (and its associated requirements) well enough to determine what goes with what, what requirements drive stand-alone functions, which functions aggregate to form lower or upper level requirements and functions, and how to reduce the amount of CI's to a logical hierarchy that provides enough, but not too many, specifications and associated documents. Perhaps it is here, in the process, where trouble with defining CI's emerges - the connections between documents and CI's is critical to understand. The challenge is to define the knee of the curve: enough selections to provide adequate visiblity to the design or development requirements without having too many documents generated during that process to drive a disproportionate number of (or "too many") or time investment in documents. Skilled CI definition means less time needed for programming, quality reviews, CM, and DM. More - or too many - CI's means more associated activity from all of those functions, causing recursive, iterative, or inefficient activities that take too long and achieve too little really value-added investment of time. The essence of good CI definition is determining which functions are not only stand-alone, but also can be tested alone and used independent of other software that is being developed - or which can be aggregated very clearly to meet higher level functions that are needed. The essence of good program or project management is to reduce the recursive, iterative, redundant, or unnecessary tasks associated with the project. Effective CI identification is the first step in the process. Cynthia C. Hauer is the Chief Executive Officer of Millennium Data Management, in Huntsville, Alabama. She has 21 years of experience in Information Technology which includes extensive involvement in CM, DM, data base design, user interface, data storage, CALS and all facets of system design and implementation. Ms Hauer holds a Bachelor of Science in Computer Science and is certified as CMIIC and CCDM, certifications in both ICM/CMII and NDIA, respectively. Cynthia is an occasional contributing editor for the CM Journal and writes a regular "rant" for the CM Basics newsletter. You can reach Ms. Hauer by email at chauer@cmcrossroads.com.
Set as favorite
Bookmark
Email this
Hits: 6480 Trackback(0)Comments (0)
|
| Last Updated on Tuesday, 24 January 2006 03:42 |
TrackBack URI for this entry
Subscribe to this comment's feed