Earlier this week I gave a talk at Bondev, Bonnier News' developer conference, about how I currently work with AI coding agents. Two workflows took up most of the slides, but the real point sat at the start and the end of the talk.
Companies pursuing multi-brand platforms know what they want but are vague about how to get there. Two principles from the world of platform providers should guide them.
For 15 years I've practiced Behavior Driven Development, focusing on what to build rather than how to build it. It made me a better developer but it also made me indifferent to most new technologies. Then AI coding agents came along and everything changed.
Last week a colleague asked me what the purpose of the async keyword was in JavaScript. Not because he didn't know how to use async/await. He was wondering why the async keyword was needed to use await. This set me off on a quest to find the answer. To avoid beating around the bush: its purpose is backwards compatibility.
JSON doesn't have a date type. Yet ElasticSearch can automatically map date fields for us. While this "just works" most of the time, it can be a good idea to help ElasticSearch help us by instead using naming conventions for dates. Here's why, and how.
There's one situation where we need to help ElasticSearch to understand the structure of our data in order to be able to query it fully - when dealing with arrays of complex objects.