Using Third-Party Code Is Not Immature

One aspect of growing in a career in development is knowing when to write your own code and when to use the work of others.

So many libraries available via others work.
So many libraries available via others work.

This isn’t a hard and fast rule, but when we start out it’s like we want to roll everything on our own. I don’t know if it’s because we want to prove to our employer, our clients, or ourselves we can do it. Maybe it’s a mix of the two or maybe we just welcome the challenge.

Maybe it’s all the above.

Whatever the case, we’ll eventually work on larger projects and eventually become more specialized. That is, we’ll identify what we need to write and where we can use an existing library.

Third-Party Code (And All The Feels)

This results in a mixed set of feelings. Sometimes I think…

  • It feels like we’re “selling out.” I mean, we’re having to incorporate someone else’s code into what we’re building.
  • We’re left wondering “could I have written this if I had the time to do it?”
  • What if the third-party library isn’t maintained and ultimately hurts our project?
  • What if the library introduces more complexity into the application than anticipated?

All are common questions. I’m sure I’m missing plenty of others, but the point I’m getting at is this:

Knowing when to roll your own solution versus using the work of others is the sign of developer maturity.

Honestly, I can’t sit here and write these words without saying that I’m not as mature as I’d like to be. But I know I’m further along than when I first got into this field.

I’m Still Immature

Just like anyone starting something new, I was out with something to prove. I felt I needed to prove to whoever would pay attention even if it was only myself.

That actually sounds kind of pathetic when I write it down, but that’s how it goes sometimes.

Thankfully, I had a group of people who helped me see things differently. I learned when to roll my own solution and when to take use the work of others.

Sure, this is a fantastic aspect of open-source software. But this was during a time when I was working in closed-source software. I mention this because the divide between the two isn’t as stark as it sometimes appears.

But I digress.

Though I go through a process of trying to understand:

  • The problem at hand
  • The problem’s domain
  • A potential solution for the problem

I also try to identify what may needs to be custom code and what can be third-party code. In doing this, I still have that pull within myself wanting to write the entire thing on my own.

Aim For Maturity

But that’s foolish and it’s immature. That’s not why people specialize if writing small utilities for unique problems. It’s not how others build businesses on top of a single, niché solution.

The deeper we get into development, the more we should be thinking through what we have to write and what we don’t. Ultimately, I believe this will contribute to more successful projects and our own maturity.

3 Replies to “Using Third-Party Code Is Not Immature”

    1. I generally agree with you, but I don’t think it hurts to peek at the code to make sure it’s well organized and documented.

      If it’s more or less something that “just works” but it does so at the sake of performance, organization, maintenance or something else like that, then is it something you really want in your project?

      Perhaps a better way of putting it is that I think there should be a basic litmus test that third-party libraries should pass before being incorporated (of course, it depends on the nature of your project and the size of the library, too ;).

  1. Hi Tom

    Agreed. There’s a time and place for everything.

    There have been projects that have required totally custom code where the only source of info has been php.net and the odd niche tutorial. Other times I’ve automatically knows someone must have done something similar before – usually many times.

    I’ve said to others in the past that there’s now a critical mass of info on the web, certainly for code, we reached it 5-6 years ago. My first instinct now is to trawl for suitable code: classes, modules, frameworks, and then go from there.

    I don’t tend to use the code as is. Invariably it’s searched through to firstly check the coding standards – easy to gauge someone proficiency. And then ripped apart / reverse engineered to suit the purpose. Sometimes it just stimulates an idea for new code.

    It cuts both ways as well. I’ve a repository of code from 15yrs of PHP development. Most of it sitting on hard drives gathering virtual dust. Some of it deserves to be left there, as testament to learning a craft. Some is deprecated & obsolete and there’s more suitable code out there. Some though is functional & either has been used as part of a project, or is still current. That’s where places like github come in.

    I’m consciously moving as much of what’s ‘potentially’ useful to online repositories, if only to keep it all in one place, with the community benefit that it’s there is anyone finds it useful.

    S

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.