NSF Research and Development Proposal (SBIR)

by Roy Roebuck

There is a common and critical need for a capability which provides Enterprise Management (i.e., human endeavor), where the enterprise, in its environment, can be seen as a whole, by all involved, during the life cycle phases of:

(note the correspondence of the above to the "plan, do, check, adjust" model of quality management.)

To enable an enterprise to be seen as a whole, and for its members and stakeholders to be supported in their awareness of the enterprise and their role in it, a networked database application for Context Management, using commonly accessible existing technology and network infrastructure, can be built and fielded in a rapid time-frame, to provide automated support for contextual awareness in devices, applications, individuals, teams, organizations, enterprises, communities, regions, nations, and the world.

By dynamically modeling and supplying information relevant to a person's evolving aggregated context, Context Management would provide a tool to reduce information overload while providing relevant data for their activities. Current world technology efforts at indexing [Rod Welch], linking [Rod Welch], profiling [timestamping aspect of text stream - Rod Welch], customizaton, privacy, categorization, search engines, data mining, data warehousing, online analytical processing (OLAP), decision support, etc. are all seeking to achieve some degree of Context Management, but are starting with a fragemented, rather than a wholistic, perspective.


Context management

... the combinations of contexts operate within a generalized "requirement" life cycle

Requirements fit within a larger life cycle for the enterprise.


Enterprise Management entails:

These multiple views during enterprise life cycle phases require new ways of structuring, working, and using data to gain true information for situational awareness, with minimal data-noise. There is more standards-based structuring of data within "object", using the OpenGroup and OMG DTMF Common Information Model (CIM) metaschema as foundation.

  1. tree pattern (object hierarchy):

    1. class (superclass/subclass)
      data-type
      attribute
    2. instance

  2. star pattern

    1. pairwise association of objects
    2. strings of pairs
    3. decomposition tree for each pair node

  3. arrow pattern

    1. change tracking
    2. change planning
    3. change informing

Enterprises are thus beginning to adopt and transition to an "object" management-data pattern. This object pattern, with its Class/Instance/Attribute, Profile/Attribute, and Change models are capable of providing a shared, distributed, and secure infrastructure, along with a coherent system "managed-object" base, which implements the OpenGroup CIM metaschema.

  1. "Tree" data pattern (Hierarchies)

    1. Computers: NIS+(Unix), FNS/XNS (Unix), NT ADS
    2. Networks: (OMG CIM, CSC SIS/SES, etc.)
    3. LDAP
    4. XML
    5. Search Engines, Indices, and Concordances
    6. etc.

  2. "Star" Data Patterns.

    Tree+Tree="Star" (single subject/entity multidimensional data),

    Star+Star="Snowflake" (single subject/entity multidimensional data with detailed depth, i.e., drilldown), and

    Snowflake+Snowflake="Snowball" (multicentric data=any subject/entity multidimensional data with depth, i.e., drill-anywhere)

    1. Associations
    2. Relations
    3. Joins
    4. Constraints
    5. Profiles
    6. Customizations
    7. Privacy
    8. Data Mining and Index Categorization
    9. Multidimensional and multicentric Online Analytical Processing
      (OLAP) for Decision Support
    10. Etc.

  3. "Arrow" data pattern (Change over Time)

    1. Change Logs

      1. Source for data Replication/Distribution/Synchronization
      2. Input to History Store (for Data Warehousing and analysis)
      3. Generated by Planning and Administration (for projecting and tracking potential and actual change)

    2. Life Cycle systems

      1. executive (strategic management) functions

        1. planning process

          1. mission
          2. vision
          3. goals

            1. performance measures (service level agreements)
            2. strategies

              1. current operations
              2. requirements and new initiatives

            3. implementation and measurement
            4. review
            5. assessment of mission value-chain (customers, products, production, suppliers) and environment (government, cultural, legal, technical, economic, competition, partners, public, political)

              1. strengths
              2. weaknesses
              3. opportunities
              4. threats

        2. analysis process
        3. decision process
        4. direction process
        5. measurement process
        6. assessment process

      2. production functions

        1. mission (line of business) processes

      3. resourcing functions

        1. information resources
        2. person resources
        3. financial resources
        4. skill-set resources
        5. materiel resources
        6. facility resources
        7. service resources
        8. time
        9. space
        10. energy resources

This Context Management application has been researched, conceptualized, designed, and prototyped in several database environments, including the latest prototype in MS Access 97. The concepts, design, and related documenation are copyrighted by Roy Roebuck, 1982-1999.

Request NSF fund our team to fully document through an appropriate standards body and implement the Context Management application as a Web database application, providing directory-centric, certified permission, automated awareness support and strategic management for the enterprise, scaling from the individual to the global. Our team will then deploy the Context Management application as an open-source, freeware application on the Internet. They will commercially benefit by serving as consultants, trainers, systems integrators, service providers, and outsourced operators of the resultant global context management environment.



Copyright Roy Roebuck, 1982-1999, Arlington, VA.