Ask HN: What technologies did you learn in 2018?

Elixir/Phoenix and React Native are my 2 main goals for 2019 :) I'm a full stack engineer and I feel like

- React native will allow me to broaden my pool of clients by being able to do mobile apps - Elixir/Phoenix and this live view might make my web dev life much easier on a lot of project by not having to use the js ecosystem!

Nothing. Strangely enough, I can't think of anything new techy that I learned this year.

I've worked full time remote in mobile and machine learning the last 3 years, but this year is special. This year has required me to set my own motivations and goals aside to really be there for my 18 months old daughter, and for my wife. At first I tried to resist and focus on both my family and my career, but I found it hard to be my best self for them.

I took a step back from the stress and greed. I learned that I need to keep my career out of my head when I'm caring for my daughter. I'm more motivated than ever to do some cool projects in 2019. And I've learned that I need to maintain a mental balance between work/passion/me and my family.

Towards new beginnings, and growing as a person.. :)

1. Do you often try to implement data structures like this? In my practice, arrays are almost always better. 2. Beside that, you may argue that I use "Don't have therefore is not required" argument. But it contained partial truth: if you fight Rust program model then you have to do some extra steps to achieve it. Rust borrowing model considers one owner who can share its data among consumers. And this is damn common situation. I written a whole telegram bot with futures, using Arc container only once, to share my http client across the threads.

If you fight GC in Java then you have same troubles: all this stuff like weak pointers, byte array storages, reflection, etc, etc... If you are trying to write imperatively in haskell... And so on. If you fight language ideas, then your code becomes more complicated. The answer here is that you shouldn't have to do it often.

LLVM. I've been learning by creating a toy programming language and compiler, implemented in Python with the llvmlite library.

https://github.com/syegulalp/Akilang

There's a couple of included demos like Conway's Game Of Life.

Sorry it's Windows only for now! I'll add Linux support once it's all not quite so in flux.

* Physics Simulation - I wrote 2D physics from scratch for this game http://noisyowl.com/disks/ and the ability to actually use the implementation of the physics engine as part of how you solve design problems was absolutely worth it. If you're a programmer making a physics game, I'd recommend writing your own physics as long as you need at most one of: gravity, polygonal objects, and rotating objects. I think I might try something using one of these engines for 7DRL this year.

* Unscented Kalman Filters - These are really cool, wow. I'm pretty sure the unscented transform has a lot more potential than just being used with Kalman filters too, someone should investigate that. Also this year I learned that the truncated Kalman Filters I've been developing for the past few years are a lot more useful and powerful than I realized, and I'm interested in exploring more general ways to use those. (Unfortunately Google won't let me publish these though.)

* Vertx server - Web server frameworks generally manage to make me mad, and this is the first one I was actually happy enough that I'll keep using it. Has a few rough edges though - I wrote a multiple database query method by hand, and Vertx's JsonObject type makes it really easy to accidentally create an infinite loop.

* Elastic beanstalk - Is actually kind of terrible? And yet it's so much closer to what I actually want (run my jar on X machines, reboot as needed, never think about it again) than what I've used before (actual servers, ec2, app engine) that I'm also sticking with this.

* Postgres - I mean I've technically used postgres before but I wrote a reddit clone from scratch this year so I know a lot more now. Hey it's as good as everyone says, go figure. Also pgadmin is so bad, wow.

* Peg.js - Parsers have always been intimidating to me and playing with this really cut down the intimidation factor. I'll write my own parsers for anything important, but still use this for prototyping grammars. The idea that you can have a dirty prototype of a grammar is mind-blowing to me though.