2025-07-20 Weekly Notes
20 July 2025
Last week I had a training event with the AI4ER people on how to prepare posters and slides for academic purposes, followed by a BBQ (and the inevitable rain). I had to present my research and I got valuable feedback from Adriano Gualandi on how to properly deliver my message and present using animations, to grab people’s attention. To be honest, I had rarely used animations before in my slides because I appreciate simplicity and with special effects it can get too much easily. However, with the feedback I got, I think the power of the animations relies heavily with the storytelling aspect of the presentation and how your speech is going. This was super valuable so I will be working on that over the summer because I know that I will have to present at my department’s PhD conference in November.
After this, I was working on the paper but I realised that my code had a minimal but very important error when it was counting the trees for all the regions in England. It took me two days to find the bug. It turns out, in the final count I was including files from different years for certain tiles (see OS national grid here). This was not only causing problems with the total tree count but also with the inequality metric I designed, so I had to re-check most of the code. Unfortunately or fortunately, depending on who you ask, LLMs are still not good enough at catching this kind of errors, and the solution was very simple. In the end, the current tree count for England is 183M trees.
Finally, and a bit unrelated to my research, in the last couple of weeks while talking to friends from all over the world (the best thing about Cambridge) and after reading the news, I realised that many former European colonies in the Americas celebrate their independence over the summer months (or winter if you go down south), so in commemoration of Colombia’s own national holiday (July 20th) I tried to plot that trend over the weekend and see if that was actually true. I posted the full version on my LinkedIn page for showboating 🤣, but here you’ll see the one that actually represents what I thought originally. Turns out, the rush for Q3 (Jul-Sep) independence happened before the 20th century for the most part (17 out of 22 countries). After that, my hypothesis doesn’t really hold true, particularly for the Caribbean countries. Purely made with R, the code can be found here.
