Skip to content. Skip to navigation Oracle Application Development & System Programmer Sacramento California
Go to TELECOMMAND HOME  
Home | Services | Contact US | Solutions | Industries | News/Events | About Us  
Document Actions

How to Become an Effective Data Modeler

by Ravi Verma last modified 2007-09-26 03:50
ravi.jpg
Ravi Verma, Chief Executive Officer

Many people call me one of the most fastidious data modelers they have ever met. I have never challenged that label because I believe that the data model for an application is as important as the foundation of a building. Inadequate attention to the model may result in irreversible errors in an application.

Those who accept my perfectionist approach often ask for guidance as to how to become a good data modeler.

I worked hard to learn data modeling and keep myself up-to-date. Here are my recommendations:

  1. Learn the rules of Normalization by heart
  2. Respect the Rules of Normalization
  3. Read good books on data modeling
  4. Find a mentor
  5. Become involved with a project
  6. When in doubt

 

Learn the rules of Normalization by heart

Many data modelers think that it the rules of Normalization are not important. Some of them actually find even talking about the rules of Normalization as a distraction from their jobs.

This is dangerous. The rules of Normalization are the foundation of relational data modeling.  While understanding and memorizing the rules is not easy, it cannot be avoided. 

Respect the Rules of Normalization

Once you have learned the rules of Normalization, it is important to respect them. Rules are sometimes tedious to follow, especially since there is usually no one enforcing them.  I have not seen many data models getting audited. The responsibility lies with the data modeler to remain true to his/her profession.

Read good books on data modeling

Make books your best friends in your quest to become a good data modeler. Once again, let me warn you that these books are not easy to read. But you must persist and read them. I recommend these resources.

Find a mentor

Finding someone who has mastered the art of data modeling may be difficult, but mentors make learning this complex task more manageable.  I might be willing to teach a class if  I can find the time and the group is large enough, but otherwise, research experts in your area and get to know them.

Become involved with a project

Nobody has been able to learn to drive by reading the DMV manual alone. Once you have learned the basics, find a project to apply and perfect your skills. Sourceforge.net may be giving you some options.

When in doubt

Last but not the least, there will be occasions where you will be lost or confused. You will find the data model too big or too complicated. You will be tempted to take short cuts. I recommend that you resist the easy way out and go to rule number#2. You must respect the rules of Normalization.

 

Telecommand Software and Services

This site conforms to the following standards: