October 28, 2004
Normalized data is for sissies
Normalized data is for sissies (kottke.org)
Something that's always confused me is the near-religion of data normalization among programmers and database admins
There are 10 [i.e. two] types of people in this world: those who work with large amounts of data, and those that don't. The real issue is data integrity, and how you manage it/create it. You'd be surprised how much data is rubbish - converting stuff into information is a damn sight easier said than done.
Managing data, particularly ensuring that it is captured appropriately and then KEPT consistent, is often ignored until things are horribly broken. Nothing causes confusion like someone using out-of date data/classifications/structures etc. because they weren't concerned with data integrity/use or just weren't paying attention.
Not that I'm dealing with that at the moment. Not at all.
