I’ve been kind of quiet in the weeks since leaving Microsoft. It’s been a good quiet - in fact, the best kind of quiet there is. It’s the quiet the results from being completely overwhelmed, engrossed, and excited about the challenge ahead. I’ll describe some of these challenges over the next few weeks in more detail, but one of them is about making large scale architectural and technology shifts. Specifically, my team is out in front, figuring out the best way to design and build lots of interoperating large-scale services that run reliably and cheaply in the public cloud. While we’re certainly open to suggestions regarding our technology choices, at present, we’re envisioning a stack based on Unix design philosophy - and as such, our current stack is Linux (currently CentOS, but really interested in CoreOS) + Docker + NodeJS + MongoDB/Couch/Postgres + a limited number of AWS services like S3.
As you can imagine, there are tons of things that we need to have on our radar to pull this off. Here’s the list that I’ve compiled so far.
- JavaScript promises library exposure (Q)
- Grunt
- Logstash
- Docker
- Vagrant
- Puppet and/or Chef
- EcmaScript 6 features - particularly iterators and generators
- Koa.js
- SSL
- Fig
- EditorConfig
- MEAN.js
- Nginx
- Memcached
- CoreOS
- Quay.io
So my question to you is this - what else should we add to this list? What’s coming that we may not be aware of? We’re working hard to build continous learning into our workday culture (not something that we expect folks to do solely in their spare time), and we’re going to be tackling many of these together as a team, so the more items to plow through, the better for the whole team.
On that note, if reading this list of research items causes you to say to yourself, “now that’s the kind of team I want to be on!”, reach out to me - we may be able to make that happen :)