Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
Home -> Community -> Usenet -> c.d.o.misc -> normalization
Say, I have a table like this:
id sw_name sw_type
1 oracle DBMS
2. windows OS 3. linux OS 4. mysql DBMS 5. tomcat web server 6. weblogic J2EE server 7. websphere J2EE server
Because sw_type contains repeating values and the normal action is to separate it into another table:
id sw_name sw_type
1 oracle 1
2. windows 2 3. linux 2 4. mysql 1 5. tomcat 3 6. weblogic 4 7. websphere 4
id sw_type
1 DBMS
2. OS 3. web server 4. J2EE server
That is fine. But, are we calling this procedure normalization? Because I found that neither 1NF(ensuring each column contains atomic value), 2NF (ensuring no non-key dependency) nor 3NF (eliminating transitive dependency) fit our action here.
If not, does that imply achieving 3NF still doesn't guarantee no redundant storage in our design Received on Sat Apr 01 2006 - 19:34:18 CST