Browse Prior Art Database

Implementation Independent Description of Data Modelling "Roles"

IP.com Disclosure Number: IPCOM000113480D
Original Publication Date: 1994-Aug-01
Included in the Prior Art Database: 2005-Mar-27
Document File: 4 page(s) / 157K

Publishing Venue

IBM

Related People

Flagg, B: AUTHOR [+4]

Abstract

A method is described to model the multiple roles (contexts) that a data object may possess in an implementation independent manner. The solution detailed is semantically consistent, capable of algorithmic translation, and is able to accurately reflect the multiple roles typical of data objects defined for real world applications.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 34% of the total text.

Implementation Independent Description of Data Modelling "Roles"

                       organization
                    _______________________
                    | orgType  | orgName  |
                    |__________|__________|
                          organization
                    _______________________
                    |      orgName        |
                    |_____________________|
                     /        |           \
 fullfills role of  /         |            \
                   /          |             \
  ______________         ______________         ______________
 | manufacturer |        |   vendor   |         |  customer  |
 |______________|        |____________|         |____________|

   Example of modelling different roles for an organization

      A method is described to model the multiple roles (contexts)
that a data object may possess in an implementation independent
manner.  The solution detailed is semantically consistent, capable of
algorithmic translation, and is able to accurately reflect the
multiple roles typical of data objects defined for real world
applications.

      A key part of the solution is the development of a consistent
set of syntax and associated semantics to express data "roles".  The
syntax is rigorously defined so the definitions can be
algorithmically mapped to any of the known data modelling
methodologies (Entity-Relationship, Object Oriented, etc.).

1.  Need for Modelling "Roles"

      In a strict and simple sense, a "role" is an optional
characteristic of a data object.  A data object is the data modelling
of a real world thing.  (Those familiar with Entity-Relationship
terminology will know a data object as an entity.)  Roles can provide
additional detail about a data object in two different manners.
Roles can be used to define differing data object types or roles can
be used to define differing contexts for a data object.  An example
of each will make the distinction.

      When roles are used to define different data object types, a
data object can assume only one of several possible roles.  For
example, a person data object can be male or female - but not both at
the same time.  In this example, the data object person would be
modeled with a type attribute of sex which can take on the values of
male or female.  These type of roles are fairly basic and simple to
model with existing data modelling techniques.

      Roles can also be used to define differing contexts for a data
object.  For example, an organization data object could have a role
of customer or manufacturer or both.  So long as the organization is
only...