When I first started my computer science degree just over a year and a half ago, some of the biggest hurdles I had to overcome involved just reading and writing code. Things like knowing what methods I can call on an array, figuring out what that error I keep getting even means, and for the longest time, why my use of recursion gave me unexpected results (hint: it’s because I didn’t actually understand recursion as well as I thought I did).

Eventually, those difficulties start becoming a thing of the past as I got more experience. But sometime around then, another set of problems start to arise: knowing how to represent real world things with data types, data structures, and combinations thereof. The first time I realized this concept was during the final project in CS162, where I had to implement a board game called Janggi. Figuring out that I can represent a game board as a 2D array was pretty significant to me at the time.

But, like the first set of problems, thinking of logical ways to represent real world things with data becomes more normal.

However, most of the difficulties I tend to face now involve understanding the subject the problem I’m working on relates to. Since my capstone project involves machine learning and audio, those are two entirely new subjects I must understand to some degree in order to complete the project. And after nearly three weeks into the project, I have written little to no code as I have spent most of my time reading documentation, reading related papers on those subjects to get more background knowledge, and designing the architecture.

If anything, writing the code is the easy part.