Transcript IS605/606: Information Systems Instructor: Dr. Boris Jukic

IS605/606: Information Systems
Instructor: Dr. Boris Jukic
Managing Information
Resources
Data vs. Information vs. Knowledge

Data: Raw (non-processed) facts that are recorded
–
–

Information:
–
–

May have an implicit meaning
May be devoid of meaning if context not provided
Processed data used for decision-making
Data provided with specific context
Knowledge
–
–
Skill, know how
Information with implied direction or intent

Intelligence (as in military or business intelligence)
Managing Data
The
ThreeLevel
Database
Model
CONCEPTUAL
LEVEL
LOGICAL
LEVEL
PHYSICAL
LEVEL
Four (Logical) Data Models

Hierarchical Model (Legacy)
–

Network Model (Legacy)
–

More than one parent allowed
Relational Model
–
–
–

Standard tree-like structure
First truly data and structurally independent model
No predetermined navigational maps as in two older models
The Database technology of choice
Object Model
–
Tables become objects
Managing Data: Getting Corporate Data
into Shape

Database administration
–

Data Administration
–
–

Using and managing DB software and hardware
Managing data architecture and definitions
Until recently, not always taken very seriously
Problem of Inconsistent Data Definitions
–
–
–
–
Same data in different files under different names with different
update cycles
Different data with same name
Inconsistent view of the facts within en organization
ERP often viewed as the best solution

Software or Policy?
Enterprise Data Planning

CASE
EXAMPLE:
Monsanto
Enterprise Data Planning: Monsanto

ERD: Enterprise Reference Data
–
–
Same set of tables used for different purposes
Single master table can be presented in many different views
(combination of columns)

–
This is in contrast with the “stovepipe” approach


Purchasing tables (databases), accounting tables (databases),
engineering tables (databases)
ERD “Stewardship”
–
–

Purchasing view, engineering view, accounting view
“Data Police” function: independent form the rest of the MIS
department, enforces data standards
Entity (Table) Specialists: key personnel most knowledgeable and
interested in particular group(s) of data: purchasing, engineering, etc.
Use “standard” external codes whenever possible
–
–
Others are likely to use them
Tested for validity and uniqueness
Four Types of Information
Data Records vs. Documents

Data records
–
–
–

Explicit structure
Defined rules
Use standard DB tools to search and query
Documents
–
–
–
Implied (or no) structure
Ambiguous rules with many exceptions
Hard to search and query with standard tools

Specialized algorithms needed
Bridging the gap between the
documents and records

Example: business letter formatted with XML
–

http://people.clarkson.edu/~bjukic/IS400/examples/ch20_XML/letter.xml
E-R Model in class
Web Content Management

Old way: Webmaster encodes a document in
in html and posts it as a file on the corporate
web server
–

Each department does it independently
New way: content is dynamic and modular
(XML)
–
–
Tags have meaning beyond formatting
More systemic approach is needed
Content Management
1.
Internal and external content
•The way content is
seen by others
•Outside-in
approach
•Localization
•Multi-channel
distribution
3.
•Content management
software
•Document as a
database
•The way content is
structured internally
2.
Case : Eastman Chemical Company


Flat HTML files create a maintenance
bottleneck
Content management product based on
preapproved templates
–

Also manages rights to update or publish web
documents
Site redesign based on external markets
rather than internal product divisions
–
See site index