Don’t tell my kids, but they’re still not very good at Pig Latin. Most dev companies get languages wrong, too.

Not that Pig Latin is an official language. To speak it, take the first sound, move it to the back of the word, and add an “ay” sound that provides a gibberish quality.

Developer becomes eveloperday. Engagement is gagementnay.

Now that my kids are deep into Spanish on Duolingo, Pig Latin is our best opportunity when we want to NOT be understood.

You might find yourself in miscommunication with developers “cidentallycay” (sorry—“accidentally”).

When I got to Chapter 7 in ​Hands-On APIs for AI and Data Science​, I was pleased to see author Ryan Day had recommended an SDK to help speak a developers’ language:

While an API can be consumed from any network-connected language, you can make integration much easier for developers with client libraries and SDKs (which stands for software development kits, but that sure makes it sound more complicated).

For many dev tools, especially APIs, a single language won’t be enough. You’ll want to support a handful of popular languages and organize ​robust documentation​ around these assets.

What I like about Ryan’s book is how well it understands the audience. His goal is to help you create APIs that data scientists will love. That field is growing and includes many first-time coders from adjacent fields. Many of these developers will use Python, which is why Ryan recommends an SDK in that language.

What languages should you support? And what does it mean to “support” a language, anyway?

You won’t be surprised to hear me say it depends upon your audience, just as it did for Ryan’s book.

What can you do to better understand your developer audience?