BETA
This is a BETA experience. You may opt-out by clicking here

More From Forbes

Edit Story

Apiary -- An API Testing Service To Reduce 'Broken' Software

Following
This article is more than 8 years old.

In an age when even your grandmother knows what an ‘app’ is, we find C-level execs and other ‘suits’ starting to understand that apps need these things called APIs.

As we have explained before on Forbes, Application Programming Interfaces (APIs) are the communications bond between different software elements and data streams. They define the route for a programmer to request services from an operating system (OS) or other application. In other words, if you don’t ‘open up your API’ (as they say), then it’ll be tough for people to connect to your application.

But APIs don’t just happen; they need to be developed and so they need a ‘platform’ (we mean a higher level grouping environment of software tools and controls) to create them in.

Apiary (sounds like aviary, but it starts with API, get it?) is a firm that sells one such API development platform. The firm’s new API testing service is supposed to allow developers to continuously verify local code changes alongside production implementation. In other words, it allows the tools and mechanics of software development to be tuned while the engine is running. The alternative to Apiary would be the manual creation of verification tests -- and they take time and money.

How to test an API

The HTTP testing framework for APIs, previously nicknamed “Dredd”, started as a sponsored open-source project on GitHub before demand from API developers led to its integration into Apiary’s wider collaborative API development platform. The product works as a testing service for users to test API specifications with the backend implementation from their local environment as well as within Continuous Integration.

Jakub Nesetril, cofounder and CEO of Apiary says that until now, “Local development testing with Continuous Integration for API specification has been a tedious multi-step process requiring knowledge of command line tools and service integration, forcing developers to write the tests manually, resulting in at least one line of code for each key. The new testing integration automatically and dynamically tests everything in API Blueprint against the implementation to ensure maximum efficacy.”

Apiary’s new testing service claims to be able to test API fidelity and reliability in live environments in real-time. As a result, API developers can ensure that APIs will work properly the first time, every time. The end result (if it works) is (in theory) less ‘broken’ software.

With the new testing support, API bugs are caught early and prevented from reaching customers, claims the firm.

Is this good news?

Well fairly obviously yes, this kind of product is generally good news for those that want it. Let's just remember that Apiary is not a charity and so while manual testing costs time and money, Apiary costs money... although perhaps less time.

Software developers will sometime struggle to create what we could call 'robust APIs' for efficient sharing of data and information between software applications -- especially given the growth of the 'API business' as a sub -sector of IT itself.

As businesses around the world increasingly rely on APIs to connect information systems, we know that a faulty, buggy API (or an improper integration) can potentially break information links between software. Expect this market to diversify and expand -- also expect APIs to be mentioned around the boardroom table.

Follow me on Twitter or LinkedIn