For me the big one was relational modeling & normalization. I took a database class in college and I thought normalization was idiotic. It seemed obviously fool

For me the big one was relational modeling & normalization. I took a database class in college and I thought normalization was idiotic. It seemed obviously foolish and wasteful and I couldn't imagine why anyone would do this to themselves by choice. Then ~2 years later I got a job working with data, at a company with quite a big database that was designed largely by experts. Within a few months of starting to work with real data I had internalized the Why of normalization and become a believer. Ironically I've been doing database work ever since. Lately I've worked for a number of companies where no one -- like seriously, absolutely zero developers on the team -- understand normalization. They may perform a shadow of it, because someone somewhere told them to, or some article said this is how you model data that looks like X, but they don't get it. Predictably, their data models are absolute garbage. Teaching them is an uphill battle because they don't care, they're not curious, and their web frameworks (cough Rails cough) have taught them to distance themselves from the database and treat it as dumb. The other one was Rich Hickey's "Simple Made Easy", which I first watched in 2013. I enjoyed it even then but I didn't really understand what he was getting it. After working with, and building, some systems done in a functional style, I feel like I figured most of the ideas. That would have been some time in 2017 so, definitely took a while to soak in.