My 2020 Tech Radar

Every year, I like to set goals for myself to help me focus on new things that I want to learn or adopt throughout the year. I typically model my goals on a layout similar to the ThoughWorks Radar that is published every quarter. The three main categories that I've used in my tech radar are:

• Exploring
  Topics listed in the "exploring" section are items that I've run into either in conference talks or papers and want to know more about them. This is usually limited to either reading a book, paper or watching conference talks on the topic. Cursory knowledge is the goal with these topics.
  
  
• Deep Dive
  Topics listed in the "deep dive" section are items that I have a strong desire to know more about and am interested in adopting into my regular practice in the future. In addition to reading up on these, I will be using the concepts and ideas in personal projects and write about in the future.
  
  
• Adopting
  Topics listed in the "adopting" section are items that influence my every day approach to designing and writing software. In addition to researching each of these topics, I also plan to mentor those on projects around me in these topics. I do find it interesting the topics I've selected for adopting this year are concepts instead of individual programming languages or frameworks.
  
  

With that general framework in mind, the goals I've come up with for 2020 are:

• Exploring: TLA+
  The idea of being able to model concurrent, distributed systems with a prover to probe for potential errors and bugs is intriguing. As backends move towards microservices, everything is becoming a concurrent, distributed system. TLA+ may never become a day to day tool but it's one I'm interested in exploring.
  
  
• Deep Dive: Rust
  I found myself writing a bit of go in 2019 on various cloud platform projects. While I enjoyed the return to more native and performant applications, go mutable by default decisions felt strange to me coming from Clojure. Rust takes the more familiar approach of immutable by default and adds compiler verifiable ownership of data that eliminates the need for a garbage collector. My gut is telling me that Rust could displace C as the preferred stack for writing apps in the future as it makes reasoning about large code bases easier.
  
  
• Deep Dive: Reason
  Reason is an OCaml dialect that the creator of React created while working on the React project for Facebook. I find the syntax easier to read than Javascript and brings along the well understood and strong type system from OCaml. I worry that moving it into adopting would be a hard sell just like Elm, PureScript and ClojureScript are.
  
  
• Adopting: Linux
  In December I took the plunge and gave up my Macbook for an Ubuntu desktop machine. It's been an interesting switch and one that's highlighted how little I know about the architecture and troubleshooting of a Linux system. I got into this pattern of ignoring the OS since I started deploying applications into cloud containers and focused more on web APIs. It was just easier to leverage some PaaS or throw away a borked image and start over. I really want to gain a better understanding of the platform that powers my desktop and the Internet as a whole.
  
  
• Adopting: Generative Testing
  This is a topic that has been working up my radar over the years. The idea behind Generative Testing is that if you can define the relationship between the inputs and outputs of a program, you can let the computer generate way more input test cases than you can think of. A set of well defined relationships, aka properties, and you can now probe your code logic for corner cases you never thought of before. Now that we have a much smaller team, I think it's time to start introducing Generative Testing and get the computer to do more work for us.
  
  

The overall theme for this year is exploring new ways of working. I just felt frustrated in 2019 when it came to supporting large, complex code based with a team of distributed employees. The focus on tooling with stronger type systems along with verification is a step in that direction. The other big theme is a return to systems programming, a topic I was fascinated with during my undergrad time. My focus on web technologies has been a solid career move but one I feel less passion for.