How to Not Suck (at Being a Programmer)

November 4th, 2013 about shells, vim, git, and regex

One of the lamest features of being a programmer is the overabundance of trivial tasks that are just plain boring and time-consuming. Fortunately, our craft lends itself very well to enabling laziness: not in a negative sense, but rather, a productivity-boosting, awesome sense. Traditionally, renaming a set of 5000 files to conform to a new naming convention (eg converting .JPG to .jpg) would take a tremendous amount of effort and an equally tremendous amount of forever. By being lazy, a programmer can expend a miniscule sliver of that effort and time to write a simple script that automates the whole process.

Update: even easier than I thought.

I'm a big fan of being lazy, so here's a bunch of tips to help you become a big fan of being lazy!

Shells

Learn to love terminals / shells and the incredible power and flexibility they offer you. A GUI cannot compete with a shell when it comes to raw power, extensibility, flexbility, or anything else that actually matters. Furthermore, there may be situations where you're confined to only a terminal and can't access a GUI. Knowing how to manipulate a shell is an incredibly understated skill in the programmer's arsenal.

Text editors

Use a real text editor. That means one that works through ssh, without a window manager. That means you need to get used to using a terminal. If you disagree with me on this point, that's fine, and I respect your opnion, but you're wrong and I hate you. GUI text editors get a few pretty features that a terminal editor simply can't support, but in the long run, they pale in comparison when it comes to features, power, and flexibility. Notice a theme here?

Regular Expressions

Regular expressions are like find-and-replace on mega steroids. I can't even count the number of times I've seen another person poking around a file, replacing a dozen function calls with a similar but slightly different function call. Not only is this a totally anti-lazy way of doing things, but it's error prone: one poor function call may go unnoticed and remain unchanged, leading to headaches for future you or future maintainers.

Version Control

This is the most important of all the things in this list. Everything else on this list is possible to argue against, but no one in their right mind would argue against using version control. Not using version control is completely imbecilic and unprofessional. Version control acts as a safety net against accidental hardware failures, file deletions, explosions, etc. Beyond that, it provides a whole host of awesome features that you should want to use anyway: easy collaboration, quick changelog summaries, project histories, systems to test new features or experiments, and more. Use version control or go home.