Zach Duda
The Cost of Halfassing
Share
Nov 15, 2022


Coding can be frustrating. But it can also be one of the most rewarding hobbies. Creating something out of nothing? Sign me up!

However, there is one fatal flaw: Each mistake you make now will exponentially multiply as you continue building your application. The best way to avoid this snowball of errors for your future self? Learn how to code properly: Right now. These little "quick fixes" are your worst enemy, especially when you already know the code running is about as botched as a haircut from a strip mall. Working code is better than broken code, but if it's only barely working then it's just barely not broken.

So how can we make our applications more stable? The first and probably most obvious solution is: USE A LIBRARY.

There is no medal for those who reinvent the wheel. If someone has a library for what you need (ie: PHP Image Resizing), there's no point in taking hours or even days in making your own. Most of these libraries have been peer-reviewed, brutally tested in production, and are usually more optimized than what you would be able to crank out yourself.

Don't be afraid to use other people's code to build your app. It doesn't make you any less of a developer! The key is HOW you use this code. If you throw the most optimized piece of software of a piece of s**t program, don't go expecting miracles to happen.

The primary reason I'm mentioning this is that I'm facing the shortcomings of my past self. Back in 2019, I had no clue how SQL or PHP worked, and I learned just this month that I could combine queries. Now, fast forward to late 2022 (nearly 2023), and this code has become a nightmare to work with. It's like trying to build a giant mansion on top of a foundation that was designed for holding a street lamp. The project in reference is none other than Empanadas.io. Both the front and back end of the website is all over the place. I tried just scrapping the current code base in its entirety and starting over until I realized: This code would take years to redo.

Empanadas.io has always been a one-man project: Me. That means that the current code base is the result of over 3 years of coding. So the cost to redo everything? I'd guess about 3 years. That is unless somehow I hired extra hands. But considering this is a parody website I made in my free time for Empanadas, I don't think we'll be turning a profit to pay for this much-needed redo.

So... what now? Well. My advice to anyone facing a similar situation: Just take baby steps. Old code is not necessarily bad code, and working code is better than broken "optimized" code. Slowly work on getting what's broken fixed, and once everything is working, bit by bit try updating individual components. I find it helpful to just made a separate directory on the web server away from production and to do all my tweaking there.

Before I go, here are some resources I wish I had used sooner: Nginx, Composer, TailwindCSS, Bootstrap, and Node.js. These programs are incredible, and it took me a while to realize just how special they are. Now... go out there and make some full-ass programs!