2016 was, as The Verge put it, “a good year for weird jazz.”1 I’d go even further: both 2015 and 2016 show that jazz is an evolving genre, and that it became more exciting than ever before. Influences of hip-hop and electronic music are becoming more visible, new artists pop-up in places you’d never expect (I’m looking at you, LA) and push music into new territories. So while I do appreciate The Verge’s recommendations (especially Shabaka and The Ancestors),2 I had to add some of my own. All of them represent that very shift in jazz’s esthetics, so if you’re looking for a review of Redman & Mehldau duo, you’ll be a bit disappointed. If you enjoy fresh sound, however, read on. Continue reading “Jazz Music in 2016”
Last weekend I spent some time working on a small project: bora.1 It’s a simple wrapper around AWS Cloudformation, so obviously everyone’s question is: why the hell would I want yet another Cloudformation wrapper? tl;dr answer is: because all the ones which are available suck. But let me elaborate.
- Troposphere-based tools are inelegant. Troposphere itself is poorly documented, and I dislike how the Python code mixes with actual Cloudformation JSON code in it. It’s also very often non-lintable (or gets unreadable after linting).
I ❤️ Python just like the next guy, but it’s not very well suited for things like CI/CD pipelines. I see this a lot in clients’ setups: first your jenkins job needs to pull the code, then create a
pipthe requirements, then lint (hah!), and then, hopefully, run. With compiled languages (and Golang especially), you only need to download a binary and run it. The only thing you have to care about is the underlying architecture and OS (which, in 99% of the CI/CD cases, is
I want to embrace Cloudformation’s new neat & clean YAML capabilities. JSON is ugly. Troposphere, as mentioned already, is ugly too. Combining YAML with some Jinja-like markup for variables and loops, we could end up with something very elegant and readable. Continue reading “bora—an AWS Cloudformation wrapper”