Table of Contents


This year felt rather quick for some reason, I feel like I barely remember anything from before summer.

Well anyway, earlier this year I set some goals to work towards throughout the year, just like the year before;

It has been quite some time since and reflecting on stuff that happens over a year is something I always find interesting. Also probably a good idea to come up and write down some goals for the upcoming year, even if I'm a few days late.

2023 Goals Recap

Learn more Rust

Last year I wrote that

"I will definitely focus most of my time in 2023 on learning more about it."

And I definitely did. The fact that rss2email was a whole year ago is crazy to me because that's what kickstarted my Rust arc.

I already mentioned the individual projects I worked on here and you can also see the repositories on my Github although a few private ones are omitted. The 2 notable ones are the ones mentioned in this and this blog posts.

I also read through a lot of articles and a couple of books on the language, both of which I learned a lot from and will definitely continue to do next year. I still have a few books I have bought but have yet to read so I need to try and stop impulsively buying new ones before those are done 😅

I even attended a few local conferences!

Build a decent TUI app

Never really came up with an idea for this, even though I always had it at the back of my mind. I guess I can say I finally did some UI at last but TUI specifically.

To be honest, I don't really mind not having done this as, again, I didn't have a good reason to at any point but I'll be keeping it in mind throughout 2024 as well. I think some sort of dashboard could be cool, I'll likely revisit this if I get around to doing anything related to distributed systems stuff as it could be useful for debugging and observing what is going on.

Learn (and apply?) distributed systems basics

Did not do quite as much as I had in mind but then again, I did not know too much about this to set any sort of accurate goals for this anyway.

I started reading "Designing Data-Intensive Applications" by Martin Kleppmann which touches on quite a few interesting and relevant topics but I still have a decent chunk of the book to finish. I also took a look and did ~half of the distributed systems challenges.

My main issue with distributed systems is that it is quite hard to find a problem that I have that could be solved by distributed systems as usually the scale required for them is way bigger than anything I am dealing with. I could build something arbitrarily like a distributed cache but I like working on things that I will actually use or at least need once.

One idea that I have for this is perhaps looking into the Actor model as that should share some of the same philosophies of distributed systems. Of course, a lot of stuff will probably not overlap but it is better than nothing.

Look into zero-copy deserialization & parsing in general

Well, I did make an ASCII Grid format parser, as well as needed to do some parsing in Shuttle's Christmas Code Hunt. The former used nom and a bunch of lifetimes that technically qualify as zero-copy. I also used pest in that project which I found easier.

I definitely feel like I could use some more practice but when is that not the case ¯\_(ツ)_/¯

Apply concurrent programming

This went quite well!

I applied a lot of stuff from the Rust Atomics and Locks book in my University software project (the one with the self driving robot), some of my solutions to Shuttle's Christmas Code Hunt (a previous iteration was quite a bit more complex but I ended up scraping it), my Mandelbrot set renderer as well as my controller ui.

I ended up using some SIMD in my Mandelbrot renderer as well which was quite interesting and I also ended up reading a lot about SIMD-able sorting algorithms for whatever reason which are quite cool!

Develop a general-purpose library for devs

I'm pretty happy with this as well. The two most notable ones have got to be my Software project and the confusables library.

In the Software project I worked on library functions that the students using our platform would use as well as a testing framework for testing some of our ROS node code that we ourselves ended up using.

As for the confusables library, I stumbled upon the fantastic talk "Plain Text" by Dylan Beattie and ended up reading up on random Unicode things, eventually coming across confusables and making a library for detecting them. I experimented with creating JavaScript bindings for the crate as well as compile-time code generation. It also ended up getting a surprising amount of downloads although I can't be sure how many of those are automated and how many are actual use cases. Still pretty cool though.

Contribute more to Open Source

This was another good one, I feel like I made a decent amount, one of them was even towards the Rust compiler! (it was definitely not a typo in the docs somewhere do not look it up)

I also made a temporary bug fix for an open source game I was playing at the time with a couple of friends. It was quite funny knowing I stopped dying to a bug that I myself had fixed once it was published.

2024 Goals

I think I want to shift my focus, at least partially, to reading more instead of building small projects. Although they are fun, I feel like I am at a point where I've used Rust extensively and doing more small projects won't teach me quite as much. I'll probably still build random stuff since they're fun and often interesting/useful but I don't think there's a point in setting that as a "goal" necessarily. I still have 3-4 books (none Rust specific) I've bought and haven't finished/started so I definitely want to be done with those and there are always more books to read if I'm done early.

I still want to learn more about distributed systems. As I mentioned earlier, I find it hard to find some tangible project idea that makes use of them to anchor myself to and build towards but reading more about them is a good starting point so I'll be doing some more of that.

Lastly, and my main goal for the year, is to look into computational fluid dynamics and attempt to build a simulation showcasing either something like the aerodynamic properties of a car or some sort of water simulation. This is quite far outside the area of things I know so I really can't tell now if the vague idea that I have of an end result is realistic to achieve over this year alone. Roughly what I have in mind is something like a wind tunnel simulation where a car is placed in the middle of the frame and air is blown out of one side, allowing you to see how it interacts with the various parts of the car. There are so many questions I need to answer even before I start doing too much research on this such as:

  • do I want this to be in 2d or 3d?
  • what will the performance look like and will I have to use a GPU? Does that affect my algorithm choice?
  • how easy do I want creating objects to test to be?
  • should I aim to support importing objects via some standard format?
  • how do I even get things to show up on screen? Should I use like a game engine or something?
  • is something like this possible in real-time or should I aim for more of a record and playback method?

This probably comes off a bit out of the blue for the type of stuff I usually do. Truth is, I've just been getting fascinated by the aerodynamics of (race)cars recently and I'm quite annoyed about the fact that I don't understand how aerodynamics work at all. Unfortunately, I was never a physics person so instead of reading a book about aerodynamics, I will instead simulate them as any non-sensible software person would do.

I would also like to blog some more this year and this whole aerodynamics arc could make for a nice little blog series.


Overall, I'm pretty happy with how the year went actually. I think I overall managed to hit or at least make good progress towards all of them. Let's see how far I get this time!

This is more or less fewer goals than last year but you know what they say, less is more and the more the merrier.

Till next year!