This is an important point: obviously, even if we don’t send any piece of data to Google, having /e/OS relying on AOSP codebase is still a dependency, because we know that Google is leading the design of AOSP even though it’s 100% open source, even though they use the Linux kernel.
But that remains true for some other open source projects. Even if you take Linux, for decades its development has been sponsored by some big techs (Intel, IBM, Red Hat…), and there is little doubt that there could have been some influence on the development roadmap and features at some point. Just look at the Linux Foundation sponsors…
But that’s the beauty of Open Source software: Freedom. Freedom to fork it if you want to fork. Go big enough and fork.
Until now, our goal was to offer a credible product that is sending no or very few data to Google, because we all know the threat that 1 single company is permanently harvesting personal data of billions users worldwide. So we found a path.
Now, ideally, we would like to break free completely from the Google, with a realistically usable product, and for this we need to go step by step.
Today, I don’t think the OS itself (AOSP) is the most problematic. The biggest concern is probably more about Applications.
Android apps rely on the Android SDK, which is fully controlled by Google, and on the Google Play Services, with two sub-problems:
- push notifications (GCM/FCM)
- the in-app purchase feature which is mandatory for many app publishers
So what can we do? We can offer alternatives to the app developers and publishers. We need good, excellent, alternatives actually. We need alternatives that offer them some benefits that go beyond the satisfaction to break free from Google because we have to face it: the majority just follow the trend and think that everything is normal.
One option that we have started to push is the support for PWAs. There is a trend for progressive web apps. They can potentially unify app development for all platforms, and it would be a good way to escape the Android SDK and any other proprietary dev SDK.
However, we would also need to have an independent push notification infrastructure. The reason is that if you take the web apps today, there is a common spec for the push API (https://www.w3.org/TR/push-api/) but you have an implementation for Chrome that is using Google FCM, one for Safari by Apple which is using APN, and one for Firefox which is using Mozilla Push Service. Not counting on Huawei which is introducing the HMS for push notifications.
So there is a road and some light, but it’s going to take some time, and meanwhile we will have to live in an hybrid world.