Towards the end of October I was already looking forward to the annual advent calendar of programming puzzles that is the December Advent of Code.
I’ve really enjoyed the past couple of years I’ve done it, especially the Part 2 twist; each day a new puzzle is presented and the first part can typically be solved with a naïve implementation, but it is often the case that there isn’t enough computing power on the desk (or likely in the world) to brute force Part 2!
Therefore it’s quite fun to anticipate where Part 2 might go, so as not to paint oneself into a corner and have to come up with a completely new approach.
The accompanying story is brilliantly entertaining, and this year for a few of the days where I didn’t have a clue where to start I popped into the friendly Reddit forum looking for tips. There I found Jonathan Paulson’s daily posts and accompanying youtube videos which were fun to watch even after I’d solved the day to see how his approach differed (not often), but he is fast - it’s like watching a speed run!
Jonathan finished 5th overall on the global leaderboard 🤩
The final day was one of those days; graph theory is something I’ve only had exposure to working with by doing these Advent of Code puzzles. I built a graph from the input data but then I had to draw out the nodes and their associations to start working out an approach:
This gave me an idea about shared neighbours, and while that did identify the 3 correct associations in the example (highlighted pink), it also pulled in 2 more 🤷♂️
I’m harbouring an intention to return to the Islands and see if I can collect the missing stars before 2024’s Advent of Code begins…
Thanks again to the creator, Eric Wastl - see you in December if not before 🤓