In
the early days of data modelling, the models were drawn up by hand, first on
paper, then using drawing tools. It proved to be a good technique for
structuring and normalising data. However, there were also some limitations:
-
Cumbersome
sharing of paper based models
-
No
embedded structured model information, such as attribute definitions, etc.
-
Impossible
to search and report on specific aspects of the model
-
Manual
implementation of data model
-
Difficulties
to keep model up-to-date
Some
of today's modelling tools have very powerful features, greatly improving
productivity and accuracy of managing your data asset:
-
Keeping
track of model changes and versions
-
Supporting
various types of models (UML, ERD, XML Schema, etc.)
-
Storing
structured embedded information for each model object
-
Generating
code for various target platforms (relational databases, XML Schema, etc.)
-
Multi-user,
concurrent modelling
-
Synchronisation
of data model and its database or XML implementation
-
Bridging
the chasm between UML and XML Schema
vocabularies
-
Web-publishing
of data models and documentation
|