Outside of a dog, a book is a man

1

• Outside of a dog, a book is a mans best friend.
• Inside of a dog its too dark to read.
• -- Groucho Marx

2
Conceptual Data Model
3
The Purpose of a Conceptual Data Model

• To understand the business
• To capture business knowledge
• It translates business requirements into data
  rules
• It defines the business system, not a computer
  system
• If done well, it will be stable across many
  systems designs

4
How Do We Do This?

• Talk to the decision makers about new
  requirements
• This is called top-down
• Look at existing systems
• This is called bottom-up design

5
Top-Down Design

• Is excellent for understanding the Big Picture
• Will help uncover new requirements
• Does not always accurately describe the current
  system
• Will be fuzzy on detail

6
Bottom-Up Design

• Will provide excellent detail of the current
  system
• Will not always identify new requirements

7
Top-Down Vs. Bottom-Up

• Shouldnt be one or the other
• Each has its own strengths that should be
  leveraged.

8
Conceptual ModelWhat it is Not

• It is not a database
• It is not a prototype
• It is not a solution

9
Design Team

• Data Architect
• End Users / Client
• Systems Analyst
• Project Sponsor

10
Data Architect

• Is responsible for the CDM
• Should work to see the CDM accurately represents
  the business
• Will need to balance the needs of multiple groups

11
End User/Client

• Is an expert in the business area
• Can find the business answers to questions
• Does not have to technologically knowledgeable

12
Systems Analyst

• Is responsible for the process side of the
  project and other needs
• Works with the Data Architect

13
Project Sponsor

• Pays the bill
• Sets the success criteria for the project
• Is responsible to upper management for the
  success of the project

14
Documentation And Communication

• Set the project scope early and keep it in mind
  at all times. Any changes to the scope will
  impact the schedule.
• Maintain the documents so that they can be shared
  among the team
• Take minutes of meetings and keep notes of every
  suggestion, request, or idea voiced by the
  customer
• Publish changes and when customer approved
• Keep a question/answer log

15
Client Approval

• Never assume the client will approve anything
• Never assume a voice approval is enough
• Show the client everything
• Get approval in writing. If verbally approved,
  document it in writing and date it

16
Client Interviews

• Who will use the data?
• How will the data be used?
• Where is the data now?
• How much is this data worth?
• Ask the questions 3 different ways?
• Dont believe them.
• They dont lie, they just dont always know the
  truth.

17
Where Can Data Exist

• Mainframe Data
• Server Data
• Enterprise level
• Department level
• Desktop Data
• Databases
• Spreadsheets
• Other
• Filling Cabinet

18
Final Document

• All the this data that is gathered is put
  together into a formal document

19
Conceptual Data Model

• is defined by the Entity-Relation Diagram (ERD)

20
Entity-Relationship Diagram

• It is graphical
• It is textual

21
Entity-Relationship Diagram

• Consists of
• Entities
• Relationships
• Keys
• Attributes
• Is not necessarily Normalized

22
What is an Entity

• In its least complex form, an entity is a person,
  place, object, or idea. It is a noun.

23
Person Entity

• Student
• Employee
• Driver
• Customer

24
Place Entity

• City
• Building
• Road
• Workstation

25
Object Entity

• Part
• Tool
• Piece of Electronics

26
Idea Entity

• Contract
• Requirements Document
• Group
• Journal of Activity

27
Find the Entities

• Read the example at the top of Page 71
• See if you can put together a list of potential
  entities

28
My List of Entities

• Customers
• Products
• Office Supplies
• Furniture
• Invoice
• Warehouse
• Contact

29
Beginning the Entity Definition

• Name
• Purpose

30
What is a Relationship?

• A relationship connects 2 or more entities. In
  its simplest form it can be thought of as an
  action verb.

31
Examples of Relationships

• USES
• TEACHES
• OWNS

32
Cardinality

• 11 Relationship
• 1n Relationship
• Mn Relationship

33
11 Relationship

• There is 1 and exactly 1 entity instance on each
  side of the relationship
• Examples
• 1 Husband is Married to 1 Wife (at a time)
• 1 Class can have 1 Instructor
• 1 Airline Passenger has 1 Seat

34
1n Relationship

• For each Entity Instance there can be 0, 1, or
  Many Entity Instances
• Examples
• 1 Class can have many Students
• 1 Class can have many Instructors
• 1 Assembly can have many Parts

35
Mn Relationship

• Many entity instances can be related to many
  other entity instances
• Examples
• Many Suppliers provide Many Parts
• Many Students can enroll in Many Classes
• True Mn are really rather rare (contrary to book
  author). It is usually advantageous to treat
  them as a 1n

36
Finding the Relationships

• Re-read the text on Page 71, this time looking
  for relationships.

37
My List of Relations

• Purchase (as opposed to Sell)
• Clients Purchase Products
• Ship
• Products are Shipped to Customer
• Have
• Stores Have Contacts

38
Beginning the relationship Defintion

• Name
• Definition
• Cardinality

39
Attributes

• Are the fields that are part of the Entity
• Simple Attributes
• Name
• Composite Attributes
• Address
• Multi-valued Attributes
• Monthly accounts

40
Attribute Definition

• Name
• Definition
• Data Type
• Numeric
• String
• Date
• Data Size

41
A Note on Names

• It will be tempting to use short names, dont.
• It is better to use long, meaningful business
  names

42
You Should Understand

• The purpose of a Conceptual Data Model
• The difference between Top-Down and Bottom-Up
  design.
• Who should be on the design team and why.
• What an entity is.
• What the different relationships are.
• What attributes are.

43
Assignment

• Read Chapters 4 5
• Understand what we have talked about tonight