API Testing, Monitoring and Debugging - A Look at the Nuts and Bolts

Wendell Santos
Oct. 24 2013, 05:30PM EDT

Continuing the theme from this morning; namely that speakers should share the mistakes and ugly truths behind their API successes, this afternoon's sessions at the API Strategy and Practice Conference gave attendees a peek under the hood. Speaking to issues that matter to API consumers, we heard nuts and bolts talks centered on API testing, monitoring and debugging.

Wayne Ariola of Parasoft, a software solutions company focused on development testing, laid out the stakes for why testing your API is critical. According to Ariola, today's switching costs associated with software are dramatically lower than at any other time making it much easier for users to try out APIs until they find one that suits their needs. Customer surveys show that 90% of customers have said that the APIs they used failed to meet their expectations while 80% have switched APIs due to failure. In short, quality matters.

To that end, subsequent speakers exposed their approaches and some of the tools to help make sure their APIs don't break. John Sheehan of Runscope started by making the point that "if you can't see it, you can't fix it" in regards to using tools to get higher visibility into API traffic. He also went over a number of useful tools for API testing shown below.
Visibility

Testing

API Science and ProgrammableWeb founder John Musser followed with a talk on API monitoring. Monitoring is the sister to testing as once developers get their API working, it's critical to have insight into how it is working, or not working out in the open. Musser highlighted the sea change happening in development summing it up by noting that "APIs are not like websites, making sure APIs are working is not the same thing." By example he showed two very similar login screens that to the end user look the same. The one that uses OAuth however has a number of API calls happening in the background. Also of note is that APIs have much more structure and semantics than the traditional web. A final takeaway was the importance of error messaging. Echoing Pamela Fox's comments from this morning, he held up the Twilio API as a standard in error messaging noting their inclusion of a URL within the error where developers can learn more about their current error.

Keep tuned to ProgrammableWeb as we continue to bring you coverage of the conference. Tomorrow we will continue with live streams of the morning and afternoon keynotes, as well as live tweets and coverage throughout the day.

Wendell Santos

Comments