Moving Away From Proprietary Software
Digital sovereignty, to me, means preserving a sense of control over my digital life. That control constantly feels under threat.
As a Gen Z man, I grew up with heavily proprietary software. It felt like there were no real alternatives for consumers. Installations seemed scary and complex, and some desktop environments looked ugly next to macOS or early 2010s Windows. Companies like Google and Microsoft offered enough concessions to appease the masses, including me. We would complain when things changed, and sometimes a decision would get rolled back in our favor. It was rare, but it happened.
I was optimistic about the internet and the free exchange of information. Growing up in a poor Silicon Valley family, I was fortunate to learn from others and build software to support myself. I hoped that initiatives like internet.org would empower the global south as it did with me, and I bombarded a close contact with questions about their work on it until they switched teams. Access to proper education, healthcare, and Maslow’s basic needs is more complex than a controlled internet portal at a local cafe or park, but I do not want to get off-topic.
Some companies have notably better taste, longer-term thinking, and less tacky decision-making than others. Lately, though, it seems every major provider of proprietary consumer software makes unpopular decisions and refuses to roll them back. I will not list examples; as a consumer, you probably have several in mind.
If I had to name the core of the problem, it is this: pushing LLM integrations without the freedom to choose your preferred model, or to easily opt out of LLM functionality entirely. Someone may want to leverage an LLM in their email or run automations on their projects, but get frustrated when it bleeds into every other facet of the product. Forcing a metric up without actually improving the UX is stupid, and eventually it degrades your user base. At a certain point, it feels like a strung-out game of telephone and a waste of computational power.
Make sure your product solves your users’ problem and is actually delightful to use.
It has gotten so bad that, at the time of writing, young executives are being praised simply for making pro-consumer decisions. Asha Sharma, the new CEO of Xbox, has received praise for listening to gamers. Hopefully the streak lasts. Meanwhile, more non-technical people are considering open-source software as an alternative.
My goal is to persuade you to move away from dependency on requisite, closed-source software whenever possible. If you choose open-source or offline apps for messaging or notes, or switch operating systems, this message succeeds. Of course, I benefit from more popular support for things I like and, potentially, some contributors for my favorite projects.
I used iOS daily and relied on Apple’s Medication Tracker, Health, Notes, iMessage, FaceTime, Photos, iCloud, and more. The experience was polished and seamless, but I was trapped in a walled garden. Improving the apps meant getting hired at Apple and hoping the decision-makers would not ruin the product. I wanted to move away from Apple for the future’s sake and to take a little more control for myself.
I bought a Pixel, installed GrapheneOS, and have not looked back. I later learned about Fairphone, which I wish I had considered earlier.
On GrapheneOS, I have fine-grained control over every permission an app requests, and I can keep certain apps contained within a special profile. This post is not particularly targeted at technical people, so I will give one example and move on: if I want to, I can go into my settings and completely disable my USB-C port, making it impossible for anyone to use the port to access my phone’s information while it is disabled.
I can finally run a browser that is not just a Safari wrapper, as everything was on iOS, with its own underlying engine and all. I use an open-source keyboard for all my needs, including integrations with onboard voice models where I select the model I want to use. I use open-source apps to log medication and weightlifting, and I hope to track all my bike rides there someday. Currently, I still use Garmin.
To give a final example (though there are many more), I have been using Signal as my primary messaging app. I love WhatsApp and use it extensively, but I do not love the data collection of my personal conversations. It has been slow, but I have been able to convince folks to come to the other side.
It is better to try alternatives and succeed than to accept current software, data collection, and imposed features. At least when a decision is made in the open-source world, you can read the discourse and understand why something has not happened. If you are curious, you can check the source code and see what is going on under the hood.
On my laptop, I run Linux Mint. It serves my daily needs and lets me work efficiently, and all my core apps are open-source. Don’t get me wrong, it is not perfect, but it is light-years ahead of Windows and macOS when it comes to finer-grained control of every little thing.
I respect Arch and Gentoo users. I used to run Arch with i3, so I understand the appeal, but I spent too much time on configurations and non-work hobbies. Now I avoid getting carried away with dotfiles. I have, though, been considering migrating from Linux Mint to LMDE to move away from Canonical software a bit.
To reinforce my point: companies often degrade software incrementally, through meetings and slow decision-making, sometimes without recognizing the drop in quality. Without open source, you cannot fork or contribute; you are left with very little recourse. Imagine not being allowed to learn how to change your own oil or swap a tire, so that even something as minor as a battery means a trip to the Toyota dealership.
Go “libre light”: swap out what you can, when you can.