the meaning of the "data" column depends on a non-key column. This would then appear to be a violation of 3NF
Ah, thank you for spelling that out. This is something I was still curious about. I have read the formal normalization rules many times, but actually seeing which applies where is still difficult for me. Seeing a problem in action, then knowing what it violates, is very helpful.
"design 3" more accurately captures the data model
Indeed, I like Design 3 more than my own Design 2. I feel kind of silly not thinking of it right away, but to my own defense, Design 2 was an evolution of what I'll now call Design 0. In Design 0, there were no aliases, only facts. When the idea of aliases came up, I proposed adding another table, and someone else proposed adding another column to the existing table. Thus, this node was born. :-)
In reply to Re^2: OT: benefits of database normalization
by revdiablo
in thread OT: benefits of database normalization
by revdiablo
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |