Software Engineering – Week the Fifth
Everything’s fun and theoretical until push comes to shove. Or maybe I mean commit comes to push. The Netflix Ratings Project was not a hard project (I’m not considering the extra credit). Basic file I/O, a little math, some loops, and some test cases as the cherry on top. Really not bad. But with school, there’s always a class or two or three or four which conspire to make little assignments tricky to schedule time for. That’s where planning comes in.
Things ended up working out this week for me: my partner for the pair programming was fantastic and we both got it knocked out the Wednesday before it was due (and I’m pretty sure it was before midnight). And that was my third programming project of the week. You might be asking yourself “How does this relate to this class or me?” It relates because this past week has shown me even the smallest tasks can become pretty tricky when the stress level increases. People always say “start early” and that’s good advice. But we didn’t exactly do most of the code early. Most of the code was done 1 day before the due date (2 if you consider the 1 day extension Downing gave because of meteorological anomalies). But we planned early. My partner and I had gotten together the previous week and planned what we were going to do. Planning: that’s the key word here.
However, planning’s utility varies from project to project. There are two types of programming problems I’ve faced: the ones where the code writes itself after you plan, and the ones where no matter how much planning and breaking down of the problem you do, you still feel like you’re a character out of Apocalypse Now when you’re trying to code–your eyes becoming flame-throwers while you try to burn a solution out of your blackened VIM window. This class seems to be based around strengthening the former more serenely-solved problem: plan, and the code will flow from your fingers like soda out of a fountain. I’d say most of the coding I did last summer at my internship was like this and that’s a really good thing. When you plan, you can design really cool structures, pipelines and layers of abstraction like filo dough–you can code code you can feel good about.
Plan, and the code will follow. *
* most of the time
2/15/2014