I’ve been writing code for about ~13 years. I’ve had jobs in the software industry since 9th grade: while my friends spent summers slaving away at their McJobs, I was in an air conditioned office browsing the web. I got my BS and masters from a good CS department, consistently ranked top 5 in the nation. Despite all that, when I got to the “real world”, I absolutely, positively, unconditionally, sucked.
Actually, I still suck, but I’m much more aware of it now and am actively doing something about it. There are many things about “professional” software engineering that you just can’t learn in school. This post will go through several hard lessons that every programmer must eventually learn or remain in perpetual suck-land forever. This list certainly isn’t comprehensive, but hopefully it’ll help some newbie programmer out there so s/he can suck just a little less when it’s time to go “pro”.
1. Maintenance
Always code as if the guy who ends up maintaining your code will be a violent psychopath who knows where you live.
Rick Osborne
The nature of being a student is that most projects are short lived: you spend a few nights on it before the assignment is due, hand it in, and never think about it again. In the professional world, it’s often the opposite. When an employer asks you to build something, remember: they are also asking you to maintain it. Possibly forever.
Just about every company is mired in “legacy” code that has been around for years and someone gets the soul-sucking work of keeping it running. Try not too laugh too much when you see it: before long, your own contributions will be part of the mess that someone else is maintaining. Once you start dealing with this crud on a daily basis, you’ll learn the true value of writing clear, understandable and maintainable code. Don’t cry too much when you come across some horrendous pile of spaghetti code, tear your hair out to understand it, curse the author and his whole family… only to come to the realization that it was something you scrawled together 6 months ago.
Pro tip: learn that programming is part engineering and part writing. You must write code that can be understood by two audiences: the computer and your fellow programmers.
2. Testing
Whenever you are tempted to type something into a print statement or a debugger expression, write it as a test instead.
Martin Fowler
Programming is hard. Really hard. Some people can’t do it at all. And of those who claim they can, they often suck at it, badly. Given how difficult it is to produce correct code, it is all the more appalling to find programmers who barely test what they write. I’ve seen people refactor classes, modify hundreds of lines of code, totally rework core pieces of logic and then check-in with no more than a cursory glance at the output. This is horrifying.
Part of the problem is that programming education is often based around manual testing: run this, click that, and if the output looks correct, you’re done. It makes sense to do this in an intro level CS course, as it’s hard to write code to test your code if you don’t know how to, er, code. However, once you get beyond this beginner stage, I don’t think any university course, coding tutorial, or book should ever encourage or accept manual testing again. For example, I’d love for every CS course to require each project to include an automated test suite: you are graded not just on the correctness of your code, but also on the (very related) issue of the effectiveness of your tests.
Pro tip: get really good at writing automated tests. They are the “healthy diet and regular exercise” of software engineering.
3. You don’t know what you don’t know
On two occasions I have been asked, "Pray, Mr. Babbage, if you put into the machine wrong figures, will the right answers come out?" In one case a member of the Upper, and in the other a member of the Lower House put this question. I am not able rightly to apprehend the kind of confusion of ideas that could provoke such a question.
Charles Babbage
I came out of school and thought I knew almost everything I needed to know to be a successful programmer. I oozed overconfidence. In reality, the gaps in my knowledge were so huge that I didn’t even know what was missing. I was not just a carpenter who only knew how to use a hammer, I didn’t even know that screw drivers, pliers, or saws existed. And I was a worse programmer for it.
Oddly enough, finishing school was just the beginning of the learning process. Every new programming language I learned, every new technology I mastered, and each new problem I solved didn’t just add a single item to my toolbelt: it exponentially increased the range of problems that I could solve and the efficiency with which I solved them.
Pro tip: you can’t predict the benefits of learning something until you actually learn it. The only way to become a great programmer is to take the plunge and try to learn just about everything you can. Read all the blogs and books you can find and get involved with open source as soon as possible.
To be continued…
Continue on to part 2 of this series to learn about screwing up, how to get things done and how to be recognized for it.
Herman van der Veer
If you enjoyed this post, you may also like my books, Hello, Startup and Terraform: Up & Running. If you need help with DevOps or infrastructure, reach out to me at Gruntwork.