Omarchy

I made the jump to Linux in 2019, and I've been using it regularly ever since.
I must admit that distro hopping is fun. I've done my fair share of it in the past. I've tried Ubuntu, Arch, Manjaro, Mabox, and Debian for extended periods, and many more just briefly.
This year I spent my August holidays at home with my wife to save money, because we are building a house. It was one of those summer days: between a bike ride and a homemade ghiacciolo, I stumbled upon an Omarchy post on Reddit. After a quick read through the manual, I decided to give it a try out of curiosity.

The premise is intriguing.
The homepage describes it as "an opinionated Linux distro". Basically, you get a ready-to-use Arch-based distro with plenty of customization already in place, some easy-to-use scripts to install optional packages only when you actually want them, and the freedom of Linux to customize it even further!
The look is very clean and cohesive, with Hyprland as the preconfigured tiling WM and a theming system that you can easily extend to almost any package with just a .toml file and a symlink.
Building a desktop like this by myself from bare-bones Arch would be a full-time job (and I would probably never get around to it LOL).

The installation was smooth.
Just a handful of questions (keyboard layout, username, password, and single or dual boot) and less than a minute of waiting.
The ISO was pretty big (~5 GB), but it doesn't require an internet connection during installation, so I find it appropriate.
I don't understand why installation wizards on other distros are often so much more confusing, especially during partitioning, where you're asked a ton of questions.
Well, actually I know why.
Omarchy asks fewer questions because it makes most of the choices for you. But that's a fair trade, because if I want a particularly intricate disk setup, I can use the advanced options. For most users, though, the installer should simply present the most common setup options.
Once you're in, that's pretty much it. You can start browsing and playing around because all the drivers work out of the box. In 2026, this is no longer unusual even on smaller distros, but it's still surprising to find this level of polish in such a young project.

After installing it, I spent a couple of weeks reading all the documentation, the design philosophy, and tweaking the system to fit my preferences.
The main things I wanted to change were the Hyprland keybindings, since I'm already used to my custom ones from dwm.
Then I had to deal with the "LazyVim situation". Neovim is the default Omarchy editor, and it comes heavily customized. Since I'm used to a quasi-vanilla Vim setup, I had to disable the autocompletion, autopairs, and autothings plugins that threw my workflow off.
I added my dotfiles, yazi, and Lite XL to complete the recipe.
All these changes were straightforward, and the new apps picked up the system theme with very little effort.

A core (marketing?) feature of Omarchy is that it provides several AI agents that you can install if you want, letting you choose which ones you prefer to use. I haven't tried them because I'm not really into AI, but I like how this is handled: since nothing is installed until I choose, it never gets in my way (apart from a few directories).
It's a far cry from Microsoft's approach, where Copilot is integrated even into the hardware.

The timing of Omarchy Quattro's release feels particularly appropriate.
Many users are frustrated with the stagnation and decline of Windows and macOS. Linux, in all its flavors, has never felt as promising as it does now, and Omarchy is a fresh alternative that deserves a spot alongside the major Linux distros.

The Omacom Foundation (the non-profit behind Omarchy) has attracted significant funding.
This is good because it gives the project more resources to develop the OS, support more hardware (like Apple's M-series and Qualcomm's Snapdragon), and sponsor some of the projects it is built on.
I have one worry, though. Many of the patrons are CEOs of companies whose services are included in Omarchy.
On Windows, even a clean installation comes with a slew of pre-installed programs (Candy Crush, I'm looking at you) that many users don't need and that take up disk space and resources. So far, everything extra in Omarchy is opt-in, but my concern is that a sponsorship model could eventually lead to something similar to Windows, forcing the project to default to sponsors' services or become bloated.

Obviously, there's an elephant in the room: DHH, the person behind the project. He's best known for Ruby on Rails, but he has also been at the center of controversy because of some of his public posts.
Personally, I believe in separating programmers from their code, just as I separate singers from their music. Kanye's controversial statements don't prevent me from listening to his songs, just as DHH's ideas don't keep me away from Omarchy.

As for who should try it, I don't think this distro is the easiest starting point for a Linux beginner. This is mainly because of the tiling workflow, which comes with a lot of keybindings to learn, and the default editor, which has its own learning curve (and many more keybindings too). But if you already have some experience with Unix-like systems, it's definitely worth a try!

I started this out of curiosity, expecting it to be just another stop in my distro-hopping journey, but I think I'll stick with it for a while and see where it goes.
With this level of funding and the plan to invest it over the next three years, I'm curious to see what comes next.

Spinning legs

I found myself getting lost in endurance sports.
When I’m climbing a mountain with my bike, or with my skis, everything feels very peaceful. My mind gets lost in thoughts, projects, ideas, and memories.
No distractions.

That’s why my favourite part of these sports is going uphill. Since I’m going slow and without the pressure to perform, I can get distracted by myself.
During the descent, even if I don’t enjoy speed, I’m too focused on not getting hurt.

I think this is another note about how much I like sports.
Sports as an act of disconnecting from society, notifications, and noise.

Mind uncluttered

Over the past ten years, I've become pretty organized when it comes to managing appointments. So I thought I'd share the simple scheduling setup I use.

First things first: I don’t trust my memory.
I don't want to use my brain as storage for things I need to remember. I'd rather keep it free for whatever I'm working on at the moment. This applies to everything, from work to hobbies and sports.

All I need to remember falls into one of three categories:

1. something that will happen (events);
2. something I need to do (tasks);
3. something I want to keep (memos and interests).

It’s worthwhile to highlight the difference between an event and a task.
Events happen even if I don’t do anything. They’re concerts, dinners out, appointments, and parties.
Tasks, on the other hand, are duties that require some type of action. Examples include paying the rent, making a call, or fixing the bike.

The events are the most important ones.
I write them on Google Calendar because I can access it even if I lose my phone. I don’t like setting a duration for them. I usually leave the default 1 hour duration, even if it can last longer, because I don’t want to add the mental effort to try to estimate the duration.
The less friction there is, the easier it is to stick to the habit.

When I was at university, I had an extra calendar for exams, lessons, etc. Now I just have a personal one.
Since I keep everything here, I use colors to distinguish between different types of events:

• orange for general events;
• yellow for trips;
• purple for work vacation;
• sparingly, other colors for special events.

I keep my tasks in Google Tasks, since it’s integrated directly into Calendar. This makes sense because tasks have deadlines.
I use brown for tasks, because I want them to stand out from the myriad of events.

I try to limit myself to 3 scheduled tasks per day. If I have more than that, I'll try to move it to an earlier or later time.
It’s just a rule that I added because I obviously do other tasks every day that I don’t plan on the Calendar, and I don’t want to overwhelm myself.

Unlike events and tasks, notes don't require a rigid structure.
I use Simplenote to keep track of interesting things I stumble upon while I’m out cycling/running or reading/watching videos.
Here I have a pinned "Quick Capture" note where I write keywords, ideas, quotes, and more. Later, I revisit those and decide whether it's worth keeping as a proper note or whether learning about it once was enough.
Less structured notes usually live forever on Simplenote. I find it perfect for me because it’s minimal, free, and syncs across all my devices.
For more structured or interconnected notes, I rely on Obsidian, but I’m still trying to figure out if it’s the right tool for me or not.
This topic probably deserves a dedicated blogpost of its own.

My system isn't perfect, but that's exactly why it works.
It keeps my brain free instead of turning it into a storage device, and it's simple enough that I can actually use consistently.

Keep it simple, stupid.

Be happy anyway

Last year, my wife and I hired a contractor to demolish and rebuild a dilapidated house we had bought.
We did the maths and took out a mortgage with a bank to cover the remaining costs. That's how the world works nowadays, I guess.
Even then, we chose not to complete the full interiors because it would be beyond our budget. I'll do the work myself or with a little help here and there in the years to come.

As soon as the work started, things didn't go very well with the contractor.
Before signing the contract, we agreed that I wanted to be included in the decision-making process and kept informed about the schedule, etc.
Nobody contacted me. I always have to be the one to call them and ask to meet, etc.

Every time we have to choose colors or things like that, we are always in a hurry because they call us with very little notice.
I even designed the inner walls because our architect wanted to draw them on site in one afternoon. It took me several evenings and like seven iterations to find a solution that I'm proud of. How could we have done that in a single afternoon?

The work is still ongoing, and I'm totally drained of energy because I live in a state of alert. I'm constantly keeping an eye on the construction to make sure that things are going in the direction we want.
All of this while I keep up with my full-time job.
I feel the pressure, but I'm willing to live with it.
A mistake left uncorrected now could be an annoying thing for years to come.

Last weekend was particularly stressful because I did some work by myself to save up, which didn't interfere with the contractor's schedule in any way. But he wasn't happy about it.
He obviously wanted to do it himself and charge us for it (it’s fair). But we can't afford those additional costs.
So, we have to adapt and postpone that work until after we move in, even if that means living with the inconvenience of a house that's still under construction.

A couple of days after that, during a stress relief run, I realized that I just have to be grateful for what I have and stop trying to finish everything before allowing myself to be happy.
I was spending the same amount of mental energy on every single detail.
It's ok to keep negotiating about the work we have agreed on, but it's not necessary to do all the work now.
A lot of families renovate their homes over time. So we can do it too when we have the opportunity.
Fine. We can deal with it.

I have to be happy.
We are going to have our dream house.
We are going to spend our lives here.
This should be a terrific accomplishment.
I want it to be like that.
And I have to work on staying positive.

Leave before locked

A year and a half after switching from Fusion to FreeCAD, I can confidently say I made the right choice.

I tried FreeCAD for the first time a long time ago, but it was too rough to get started with.
I’m used to commercial CAD software like Inventor, SolidWorks, and Solid Edge. They are all reasonably polished and stable. Back then, FreeCAD had a messy interface and wasn’t very stable.

That’s why, when Autodesk released Fusion with a free account for non-commercial use, I was happy to see a premium-level CAD software become available to hobbyists.
That excitement didn’t last long because I realized it had some important drawbacks: you must have an internet connection, all your files are stored on Autodesk’s servers, limits on the number of parts you can open at the same time, and opening a STEP or STL file takes forever because it has to be uploaded first.
Still, it was much more stable than FreeCAD, so I kept using it.

Until I heard that FreeCAD had released the holy version 1.0.

During the Christmas holidays of 2024, I downloaded it and gave it a shot.
The program had improved so much since the last time I tried it. The interface was still overwhelming at first glance; there were too many buttons and toolbars visible by default. The workflow felt strict about the order of clicks when creating constraints or performing cuts and extrusions, but most of the crashes had disappeared.
To me, this was the most important improvement!

Once I realized the core experience had become stable, I started customizing the UI and the shortcuts with the goal of making it my main CAD software because I like the open-source philosophy, the freedom it gives, and how extensible it is.

Luckily, after some experimentation, I found Deltahedra, a French YouTube channel that makes some of the best tutorials about FreeCAD.
I learned a lot there, and I thought it would be worthwhile to share the channel and the best videos and tips to get started:

• UI customization
• Beginner tutorial
• FCinfo macro wiki for manual installation (if Addon manager doesn't works)
• Basic Workbenches for beginners: Sketcher, Part Design, Assembly, TechDraw
• Edit downloaded STL tutorial (3rd chapter)
• Extra Workbenches to edit STL: Mesh, Part

FreeCAD is currently on version 1.1.
The improvements from 1.0 are significant, and they make onboarding even easier.
I even installed it on my dad’s PC because I now feel it’s mature enough to be used not only by hackers and enthusiasts, but also by casual designers.
If you are thinking about trying FreeCAD and have questions or run into problems, feel free to contact me. I’ll be happy to help if I can.

I hope more people will consider switching to community-driven software.
A change you choose for yourself may feel difficult at first, but it is still easier to accept and commit to than one imposed on you.

Open, less so

Over the past week, many Android FOSS apps installed from alternative stores have started displaying an alert saying something like:

"Google has announced that, starting in 2026/2027, all apps on certified Android devices will require the developer to submit personal identity details directly to Google.
Since the developers of this app do not agree to this requirement, this app will no longer work on certified Android devices after that time.
~FreeDroidWarn"

This effectively means that apps whose developers do not comply with these requirements may be blocked or restricted on certified Android devices. This is actually something we have known for a while, and it’s going to start in September in a few selected countries, then globally from 2027.

The issue is not only about developers.
Many independent and open-source developers may decide not to disclose their personal identity to Google, which could lead them to stop publishing or maintaining their apps.
As a result, end users could gradually lose access to a significant part of the Android ecosystem that exists outside the official stores.

For users like me, who have relied on Android's openness for years, this change feels particularly concerning.
I have always liked the possibility of installing all my favourite custom apps from F-Droid or alternative app stores.
What worries me most is not the fate of commercial apps. Those developers will simply comply with the new requirements.
The apps most at risk are the small independent projects maintained by volunteers, which are often the reason many users choose Android over more restrictive platforms.

There are still some possible workarounds, although they are not always practical.
Sadly, I can’t install an independent Android OS like GrapheneOS because, without Google services, many banking or digital identity apps would not work properly.

Since switching away from Google's ecosystem is not a practical option for me, I decided to support advocacy efforts instead.
I signed the Change.org petition, filled out the Google developer verification requirement form, and emailed the EU DMA team.

If you are also interested in this topic, I suggest reading the manifesto and taking action as well.
This is something I care about, and it’s not about Android versus Apple. The gap between the two ecosystems seems to be shrinking. I just hope platforms move toward more openness, not less.
I remember projects like Cydia showing that users do value that kind of freedom.

Back to it

The more days I go without writing, the less likely I am to write in the future.

Last summer was rough, and I broke my writing streak.
After that, I started managing the construction of my house. One thing led to another, and I stopped publishing notes.

I needed to create something so badly that, during the winter holidays, I started and finished three whole projects!
But after the holidays, work resumed, and I got busy again, so I drifted away from writing.

I never aimed to write every day, but I want to keep my brain trained to generate output that I’m proud of.
Until I have just a rough thought or a draft idea, it seems so clear. But when I try to share it, I start to see (a lot of) gaps in my thinking, and this forces me to work on it more.
This is good. I like to take my time to think about things.
The problems arise when I try to write a blog post: after writing the basic idea, I keep polishing it forever. I have a dozen unfinished notes that talk about nothing, without a clear point.
Like this one.

This may look like a meta-note, but I wanted something to break the habit and just deliver a message.
Maybe there are other bloggers stuck like me who, by reading these imperfect posts, will start writing again.

On engineering

I’m writing this note off the cuff because I just watched the latest Veritasium video, and I’m overexcited. It’s one of those moments where something you already knew clicks suddenly.

The video talks about the engineering behind ASML’s EUV lithography machines, one of the most complex devices ever built. Since these machines are responsible for producing the core of modern PCs, they are a central asset in our economy. Billions are poured into their development, a process that spans decades.
What makes this kind of engineering fascinating to me is not just the final machine, but the sheer amount of time, failed attempts, and belief required before anything actually works.

The video is very well made and manages to explain, in a way that’s easy to follow, the stages of developing and bringing something entirely new into existence. It’s an achievement that, on a smaller scale, every engineer eventually has to face.

I often find myself, when talking with friends or colleagues from other departments, trying to explain problems I’ve encountered at work. More often than not, quick solutions are offered (and I know I do the same when someone explains a problem outside my own context).
Without the full window of constraints and history, complexity is easy to underestimate.
From the outside, problems often look simple. From the inside, they rarely are.

Decisions made in the past shape the ones we can make today, yet that context is usually invisible to anyone who wasn’t there when those choices were originally made.
Of course, this isn’t true for every field or every role, but much of day-to-day work tends to focus on the present and on relatively short time horizons. You solve this task this morning, this problem today, this issue this week. Sometimes you get bigger challenges that keep you busy for longer, but they are still framed in relatively short cycles.

Engineering is different.
You constantly work on projects that span months or even years, where the choices you make today directly affect weeks of future progress. And when those choices turn out to be wrong, you often have to step back, question your assumptions, and accept delays as part of the process.

That’s what I like about my job.