JACOB HERIC


I PREFER NOT TO
What makes a good developer?
December 31, 2009

I'm not going to offer anything groundbreaking here. But having worked with a variety of teams of talent over the years, I wanted to articulate what I think makes a good Java developer. By putting these thoughts down, I hope to forge a little more of a formal idea of what I'm looking for the next time I run into new talent and have to assess it and work with it.

Unfortunately, the reality is you're never really going to know what you've got until you're working with a developer and you see him or her make that first critical decision in the project pressure cooker. Interviews help and code samples are great, but they are pretty highly controlled artifacts and evidence. They're in vitro, not in vivo. Real life conditions for a developer are fractious. Deadlines loom, clients transmogrify, PMs hound and every developer action gets instantly tabulated in the ledger of present and future cost (how much technical debt did you just incur when made that lousy decision?).

Here are my few short and broad characteristics of good devs. I'll follow up with a little discussion of how to spot these traits and why they are important. It might not be possible to know for sure you've got a good developer until you've worked with them, but it's imperative that you know before you're tearing out a third of your code base a month before your deadline.

A good Java developer...

- Has varied experience kept at an arm's length.</li>
- Has curiosity plus adaptability.</li>
- Communicates persuasively.</li>
- Prefers simplicity to novelty.</li>

Experience is essential. How much? 5 years? 10? I'll take a dev with five years of experience on five substantially different projects over a dev with 10 years of experience working in one environment. I want a developer that has strong ideas about the right way to do things, not a strong opinion about a particular way of doing things. A developer that looks at a problem and makes a decision on the best way to solve it is usually a developer who has solved it a few different ways in the past. A developer who sees a problem and immediately grabs a chunk of code and says "here's how we did that" is going to put a lot of code in your project that is going to get ripped out. Who does that? A developer who has logged 16,000 hours JDBC programming, for example, is going to do that because they have long ago stopped thinking about how to get data from the database (in order to preserve their sanity). A developer who has implemented a couple different ORM providers, used Spring's JDBC template and done straight JDBC programming (in reverse chronological order) is much more likely to have ideas about how to do data access in their tool belt rather than specific data access implementations.

The later developer with more diverse experience is also going to carry a healthy skepticism about past implementations. They are going to keep those at arms length because they've had the opportunity to compare them to different approaches which inevitably highlights shortcomings.

Which brings us to curiosity and adaptability. Varied experience sets the stage for curiosity. If you've done something three different ways you're much more likely to be curious and looking to find a fourth way that encompasses the best of the previous three and adds the missing salt. You're also much more likely to find and consider radically different or breakthrough approaches (which carry the potential for huge time savings or jumps in profitability). Have you done years of SQL, HQL & EJB QL? Then you're probably going know when to try NoSQL! Have you done JSF and used half dozen Javascript AJAX libraries? Then you're probably going to know when it's time to try FLEX.

But you can't switch to a new technology unless you're able to articulate (especially in writing) specific advantages. This almost always boil down to an effective outline of short or long term cost and time savings. And, you need to be able to do this to a variety of audiences: non-technical clients, managers, PMs and your dev peers. Being able to write a persuasive argument doesn't just demonstrates a certain quality or refinement of thinking, it demonstrates an ability to know and listen to an audience. A good developer doesn't just know their audience, he or she is sensitive to their interests and capable of tailoring communications to them. You can start to feel developers out on this front from the first email resume submission. Look for emails and dialogue that demonstrate even and measured assessment of technology rather than singular endorsements based on recent personal experience. An average dev will tell you that they've built an app using Ext GWT, a better dev will offer an opinion about when it's appropriate to use Ext GWT and probably have a short list of pros and cons to offer off the top of their head.

The last and most important attribute of a good Java developer (or any developer for that matter) is a proclivity for simplicity. Simple code is maintainable code. Maintainable code is good code. You can usually see this in code samples. Are methods discrete and short as possible? Is the formatting and syntax consistent? Does the code tell you clearly what it's doing and do the comments tell you consistently why? You can spot the habits of good developers in the simplest of things. Does code use the Java 5 for-each construct here and an iterator there? Is the StringUtils being used in one spot and a custom algorithm in another? That's a dev who's discovered a new way of doing something and switched to it without switching it everywhere. That's also a maintenance problem. Simplicity with an eye towards maintainability is an especially important attribute for a developer who does your framework selection. A good developer won't put a framework into the mix until they know how to use it and are confident that there are time and maintenance advantages.

If you talk to a developer, look at their resume, check out their code and work with them even just a little bit, you'll see pretty quickly whether or not they possess or have the potential to possess these attributes. If they do, you've got a good developer or a good developer in the making on your hands.