Why you should write good code

I have seen a lot of bad code. Disorganized, repetitive, poorly-commented, inefficient bad code. And I'll admit – I have written some bad code, too. As a researcher I understand that the incentives on how to spend your time often run counter to your best intentions, and that sometimes bad code is unavoidable.

But I am confident that most of the time writing good code is not only possible, it's also well worth the effort. Clean, well-written, well-maintained code actually saves you time, reduces errors, and makes it easier to share your code with other people. 

But perhaps you think this doesn't apply to you... perhaps you really don't have time to write good code, or your shoddy code at least isn't hurting anyone? 

Perhaps. But perhaps not. In this post I'll examine four common excuses (ahem, reasons) that I've heard for why programmers write bad code, and by the end I hope you'll agree that in fact you really can't afford not to write good code.

Reason #1: "I don't have time"

Not having enough time is by far the most common reason I hear for not writing good code. And it's true – good code often demands a higher initial cost of time and effort. But this kind of thinking is very short-sighted; in the long term, poorly written code will actually weigh you down like a load of bricks.

Poorly-written code often runs slower, which costs anywhere from a few seconds to a few minutes every time you run your program. It's also more difficult to read and understand, which means more time debugging when (not if!) things go wrong. Hastily written code usually only works for a specific case and must be manually tweaked for every variation. This kind of rigid structure is very difficult to modify and repurpose. And finally, returning to your code after an absence is almost always confusing and frustrating, and discourages you from reusing your code in the future. To summarize:

Reason #2: "My code may be messy, but it works just fine"

When I hear this reason I wonder whether this person has ever seen another person's code. Software bugs are so ubiquitous they have even been found in NASA's space explorations, Intel's pentium chips, and United States missiles and military aircraft. Even for well-written code, it's almost certain that if the code is long enough it has at least one bug in it. 

How confident are you that your messy, hastily written code, which only you have reviewed and tested, has no bugs? How many programs have you ever written in a hurry? What are the odds that there are no bugs in any of them? Pretty slim, I'm guessing. 

Writing good code not only helps prevent mistakes in the first place, it also makes it easier to find the mistakes later.

Reason #3: "I'm only going to use this code once"

OK, I will partially concede that this is a reasonable excuse for not writing great code. But stop and consider for a moment how often this is really true. How often do you return to that code you said you would only use once? Even if you don't use the entire script or function verbatim, how often do you copy sections of code to use elsewhere? 

Is there a better alternative to writing single-use code and cannibalizing it? Perhaps you could write a smaller, cleaner, more abstract function, and turn that single-use code into multi-use code. 

Reason #4: "I'm the only one who ever uses my code"

While you may be the only one using your code now, it's worth considering whether anyone else might inherit your code in the future. If you are part of an organization or group, what are the chances that some poor soul will take over your project one day and have to sift through your incomprehensible code? This has happened to me, and let me tell you – it sucks!

And even if you really are the only one using your code, I hope you'll agree that you are actually pretty important. As I mentioned in reasons 1 and 2, writing good code saves you time and reduces the number of errors you make. After all, "Your most important collaborator is your future self." And hey – if your code is clean and easy to use, maybe you won't be the only one using it for very long. Developing good programming habits opens up doors to collaboration and contribution to larger projects in the future. 

So, is it worth it to write good code? If you consider the time saved, errors prevented, code reused, and opportunities for collaboration, the benefits seem to greatly outweigh the initial costs. I hope this post inspires you to improve your code habits! If you're looking for more information on how to write better code, I'll be writing a series on best programming practices that's geared toward young researchers. Stay tuned!