Looking Back at 2025

I enjoyed making a post last year to reflect upon what’s keeping me busy, or what caused me to not be busy, so I was thinking to do it again! If you’re reading this, it turns out I went through with it.

In general, the past year has been much better to me than the one before, at least on an emotional level. The relationship has progressed into becoming an engagement, and the work I put into my home in 2024 has paid off. I’ve had more time and energy to spend on tech projects, and so I did.

I’ve started using a kanban board¹ for personal projects, so I can get a more clear overview on what I need to do, and in some cases, what order to do them in. This has helped me a lot in keeping my priorities in check, and to get potential new ideas of my mind by writing them down so I can focus on more useful things first. There’s also some positive reinforcement by checking the board once a week and seeing how much has been accomplished.

My Homelab

The biggest project I’ve been working on is a mix of Ansible and Terraform, to maintain my personal cluster². While I used to use Bashtard for this, as noted in an older blog post, I’ve reconsidered this approach. I made Bashtard as an “how hard can it be” kind of project, and the result is that its not necesarily hard to make your own configuration management system. It is, however, very time consuming. Ansible comes with most of the stuff I want out of the box. I do dislike that I cannot deep merge configuration files, but that can be solved by just making less pretty configuration keys.

Ansible manages all my physical machines, servers, workstations, laptops, HTPCs. On a subset on machines it will also configure Kubernetes, specifically, k3s. The Kubernetes deployments in turn are managed by OpenTofu. I hope to make a future blog post soon-ish going into more detail on how and why I did the Terraform resources in the way I did. For now, just accept that its been working reasonably well.

DashLab

Another project I’ve had fun with was DashLab, a dashboard for homelabs, written in Golang. I don’t do much general programming anymore, as I don’t really feel in touch with most languages anymore. Perl is always good, but it does appear to be thoroughly disliked by “modern” developers. Raku is still slow. Rust is more a cult than a programming language community, and I’ve had enough interaction with cults in my life. Python I sadly have to use for work, but it is hard to find a worse-designed language these days.

Go is not perfect either; it has strong ties with Google, and I strongly dislike having to jump through hoops to keep the Google shitware out of my way³. gccgo exists, but it lacks behind so far that it is not practically usable for me right now.

That said, the language itself is clean, it compiles very quickly, and the resource usage of running programs seems to be more than acceptable. Its easy to pick up, and feature-rich enough to write out decent looking code.

Projects for This Year

Of course, the new year comes with its own projects. Blogging more about my homelab set up is a part of it, but I want to do more than just write about things I’ve already done.

Trying out Wayland Yet Again

As I do each year, I do want to revisit Wayland yet again. Hyprland was starting to get very acceptable last year, and I hear/read it’s only getting better. Maybe 2026 will finally be the year of a stable Wayland setup with a tiling WM!

Mapping Surveillance in OSM

A more interesting project, to me at least, is that I’ve set out to map surveillance cameras in the city I live in. After being inspired by a Dutch news article, I was wondering how my own neighbourhood was doing on surveillance, but there was no data in my entire city. Since anyone can contribute to OSM, I thought to myself I might as well be the one doing it. It gets me out of the house a bit more, and incentives me to see more of the city I otherwise would ignore. I’m not sure if I’ll continue with the entire city, but I would like to at least do a large part of it.

Releasing Terraform Modules

With the work I’ve been doing on my homelab, I’ve also created a few Terraform modules which I use in my own setup. They are set up to make certain repetitive tasks easier, for me at least. They’ve been used liberally in my own setup, and I am starting to get confident that they are becoming rather stable. I still need to document them properly, and figure out if there’s some easy means of sharing them in the Terraform/OpenTofu community that doesn’t require registering on a proprietary platform. If there isn’t, then that’s fine too and I’ll just release them as simple git repositories.

Wrapping Up

As noted before, I’m entering this new year with a positive attitude, and I hope it will keep up. I’m already pretty happy with TyilNET in its current state, everything I want running is running just fine, and deploying new applications is a breeze under the current setup. I’m looking forward to blogging more about it, and maybe write some new software now that it’s so easy to get it running the way I want it.