> "Once, if you had lots of data, you had to store it in fully normalised form, because storage was expensive."
> I don't think normalization was ever about storage space.
I'm not the GP, but...
I had to sell the "extra" work and complexity to a nontechnical stakeholder who considers themselves a power user of Access and Excel and wants every report to be a single simple query with no joins that they can write and run on their own.
BTW, the same stakeholder also wanted to shard data by month through table-naming conventions (ie. SELECT SUM(Total) June_Total FROM June_Sales) and get aggregate reports by looping through these tables.
So, yeah, the winning argument came down to money for storage (larger hard drives & bigger backups of the duplicated data) vs. having an employee run reports for him.
> I don't think normalization was ever about storage space.
I'm not the GP, but...
I had to sell the "extra" work and complexity to a nontechnical stakeholder who considers themselves a power user of Access and Excel and wants every report to be a single simple query with no joins that they can write and run on their own.
BTW, the same stakeholder also wanted to shard data by month through table-naming conventions (ie. SELECT SUM(Total) June_Total FROM June_Sales) and get aggregate reports by looping through these tables.
So, yeah, the winning argument came down to money for storage (larger hard drives & bigger backups of the duplicated data) vs. having an employee run reports for him.