As another busy week comes to an end, we bring you the news we couldn’t cover with a look at what was going on in the world of APIs. This week saw a lot of news coming out of Amazon and leading off is the Amazon Web Services launch of its Greengrass Internet of Things service. The solution lets customers collect and process data from field-based devices, deploy and run Lambda applications on those devices, and secures communications between devices and the cloud. Along with the launch of the service, Amazon is also providing an SDK that allows for local invocation of Lambda functions, publishing messages and more. The release of Greengrass makes Amazon among the first of the major cloud providers to offer open first-party services for edge processing, beating Microsoft’s Azure IoT Edge and Google Cloud IoT Core to general availability.
Another announcement out of Seattle was that Amazon API Gateway now enables customization of error responses. Developers can now modify various parts of the generated response including HTTP status code, header, or response body. Previously you had to code client-side logic to process default responses returned by API Gateway. Now those responses can be customized in a way that makes them more useful to developers consuming your APIs. We’ve pointed out best practices for ensuring that your API returns meaningful error codes and this is a step by Amazon to make this even easier to accomplish.
Image recognition services continue to be refined as the sophistication of machine learning and artificial intelligence improve. On that note Amazon announced that their Rekognition service has now added celebrity recognition to its list of capabilities. Users can call the RecognizeCelebrities function to access this feature, responses will look like below
"Id": "3Ir0du6", "MatchConfidence": 97, "Name": "Jeff Bezos", "Urls": [ "www.imdb.com/name/nm1757263" ]
The Rekognition API is currently pulling information from IMDB though that may change in the future.