On theory, practice, and Snowflake Schemas

Just the other day I learnt that the data warehouse I was working on was designed using a Star and Snowflake schema. I’d known enough about them to know that this meant the data was set up on fact and “dimensional” tables, but not much other than that.

So the moment I had some time I went online and looked up definitions, and realised that they were pretty much the way many of the bigger databases I’d done up looked like. I’d been using this schema for the past four years (at least) without my ever realising it. It was like in Le Bourgeois gentilhomme when Jourdain remarks

Good heavens! For more than forty years I have been speaking prose without knowing it.

Which reminded me of something else I read about in Taleb‘s Antifragile: that academia often comes after practice. You can do something your whole life (practice), have it labelled and described as something in theory (academia), and after a while forget that it’d started not as a theory but as an unlabelled, undescribed bit of practice and not as a theory.

Personally, Taleb’s spelling this out gives me much reassurance that we don’t always have to understand something theoretically in order to do something (an activity) well. If I want to run well, or swim well, or program well, it doesn’t necessarily have to follow on from learning the theories of aerodynamics, water viscosity, or binary.

Published by

Donn

My name is Donn, and you’ll find me working at the intersection of business and information technology, constantly looking for ways to apply IT to business and life to make things better. I’m a big fan of data analysis and its subsequent communication. It always gives me a thrill extracting meaning out of data through analysis, and figuring out the best way to present the findings for maximum impact!

Let me know what you think