Databases Reference
In-Depth Information
Consequently, it is necessary to find an indirect way to
validate the data models, using prototyping in the form of a
user interface to which business users can relate. To get to
this point, a Model-driven MDM tool is unavoidable. All the
data management screens are automatically generated from
the data model. There should not be any heavy software
development needed between the moment that the data
model is available and the time that the screens (which can
be understood by business users) are available. So one could
say that the approach at our disposal is one of “
What you
model is what you get
”
.
By handling data using the data
governance functions brought by the MDM system, the
underlying data model can be more easily validated by the
business users. This model-driven approach is mandatory to
involve business users in the validation of data models.
Without this tool-based strategy for validation, business
users cannot be owners of the data models or the data
repositories on which they are based.
7.1.4.4.
What are the connections between this organization
and the IT department?
In order to succeed in data modeling with a real
participation of business users, the IT department needs to
be involved at several different levels. First, IT must put
forward the semantic data modeling procedures as well as
the tools to support the process (data modeling workshops,
backup and security of the data models). Following on from
this, IT must accompany business users and the Enterprise
Data Office so that they can adopt the data modeling
procedures and also so that they can use the MDM system as
a tool for prototyping at the time of the data models
validation.
In their own department, IT experts must translate the
data models into the technical terms expected by the MDM
software. IT architects have to handle the integration of the
