For years I have been in the Linux-on-the-desktop camp because I find it superior as a development environment. Both because of the tooling (high quality shells such as bash, and integrated package managers like apt and yum), and because of the productivity that comes from having a development environment similar to the target production environment (usually Linux in my experience, especially with the advent of cloud computing). This position was only reinforced by the rise of Docker and containerisation which are primarily Linux based.

Back in 2016, I made a new year’s resolution to write at least one blog post a month. Like many resolutions it didn’t see out the year, stalling in August. In the 4 years since, this blog has remained neglected. In that period of time, I have changed job, changed country and added a couple of children, so the lack of activity perhaps somewhat understandable. At any rate, I have resolved to start blogging again, but without setting specific targets which can only lead to disappointment when they’re not met.

Spark 2.0 introduces a new mode of computation for Spark, Structured Streaming. This post gives a very basic fully working example of using Structured Streaming. The full code is available here on github.

This is the second part of a series of posts looking at how features in Java 8 can help Scala developers stick to a functional style if they find themselves needing to develop Java code. The first part was about handling immutability.

Early on in my experience programming with Akka, I came across the post “Attention: Seq Is Not Immutable”. From that point on, I’ve always been careful to use scala.collection.immutable.Seq rather than scala.collection.Seq when designing messages to pass between actors, and in general to ensure messages are immutable.