Hi, I'm

Jonathan Pike 👋

I'm a software developer in St. Albert, Alberta.


Programming As Data Management

January 20, 2016 # permalink

As I’ve learned more about programming, I’ve gone through several distinct mental models about what I’m doing. A rough timeline:

  1. Beginner Beginner: Programming as a black box. I type a few magic words and out comes math. Awesome!
  2. Intermediate Beginner: Programming as syntax. I no longer type incantations to the mighty computer, but syntax that is parsed by an interpreter. Math still comes out.
  3. Advanced Beginner: Programming as data management.

This last mental model formed after researching web APIs. What is an API? As an acronym, I knew that API meant Application Programming Interface, but what does that mean? I didn’t know. What did it mean that Twitter had an API? What did that do for me?

Surprise, surprise – Wikipedia has the answer. An API, when it comes to web applications, means that a programmer has an interface to access the data in that app. In the example of Twitter’s API, I have access the timelines, tweets, followers, and a lot of other information, of any user that grants me access (via API keys).

In that regard, I believe can be largely seen as data management and manipulation, both in your own app and in any app that you interface with. Rendering a page? Data management. Accepting user input? Data management. Accessing a 3rd party app’s API? Data management. Any problem that you’re solving is essentially one of data management: where is the data coming from, what type of data am I getting, what do I need to do with it, and where do I store it?

Breaking the entire field down into these four (and potentially more) questions really simplifies things for me. A new idea that I’m working on requires me to fetch RSS feeds from blogs. Prior to this realization, I would just be going about trying to find a gem or library that deals with RSS and not worry about what is actually happening. Now, I know that the data is coming from RSS feeds from other blogs, the data type is structured XML, and I know what I’m going to do with it. I don’t (yet) know how to write my own XML parser (that’s another topic altogether), so I still have to rely on some sort of external library to do the parsing for me, but I have a slightly deeper understanding of what is going to happen under the hood, so to speak.

The next time you come across a programming problem, try thinking about it in terms of the data.