I am a little sick of the seemingly lost focus on existing products. As a developer myself, I understand that software development takes time and is resource-intensive, especially if what you are working on is unique and has never been done before. However, the focus of your available resources seems to not be set right, in my opinion. As a result, I will list a few things that I absolutely cannot stand about the way that Proton creates products in the hopes that I, along with many others, can show the weight of this reoccurring issue.
Starting with Linux systems. Yes, I am aware that the smallest group of customers uses a Linux desktop. This is not an excuse to have basically no decent Software support for it. The only clients even usable currently on Linux are the web clients (duh), the bridge, and a very unstable version of Proton VPN, along with its future self, the alpha or beta release. Now, I have to say here that Proton does seem to shift more focus over to Linux, which is quite nice to see. The VPN client alpha is among the best working, far better than IVPN or Mullvad, with an obvious lack of features.
As a whole, I don’t think that this is the biggest issue anyway. People using Linux are usually quite advanced users and are capable of using a WireGuard config file, which is often the better choice anyway.
In my eyes, the largest issue, which I have noticed just by reading posts a couple of years old, is that some features are planned on a platform (for parity or accessibility) and will then be followed up with mostly nothing. To me as a customer, this seems like bad management. Features in the iOS app should not take two years to be implemented in the Android client, and vice versa. You release more products and then largely ignore pressing issues. The reason I choose Proton services is not because I want a new Password manager that is replaceable by three other technically equally good or better alternatives that have zero issues, but because I want to replace services that have no good privacy-oriented alternative. Like the email service you guys got famous for in the privacy community.
If improvement of the portfolio is so important, maybe instead of pumping out products, the mobile apps should be at least accessible for the rather large number of users that do not have access to Google services out of their own choice (of which I am not one). Maybe two mobile clients should not get the same feature in the span of a few years. This example is not an issue of lacking resources. You have a rather large team working on different projects for the same platform. On Android, implementing push notification support without Google’s APIs should be rather straight-forward, especially since you have already done so in the VPN application. Yet, for several years now, there has been no update for it. This simple change would make the mail client viable for the f-droid repo. There are open-source projects that are maintained by hundreds of people at a time. You should shift your focus, and I believe you would likely succeed in creating actual finished products by focusing on one product at a time.
All of this being said, I still want to thank you. The reason for all this inconsistency across products and platforms is likely due to the high complexity of said products; I am aware of that. There is simply no other company that does it quite like you guys do. All I know is that despite the issues I face every day, I gladly try to use your products when they do work. I would not want it any other way.
Your services are hard to use (not from a UI/UX point of view), and I hope you can work on this incredible base you guys have built. The few issues I have listed here are not the only ones on my list, but they aren’t new to you guys either, so I will shorten the text wall.
I don’t have any issues with using Proton bridge and VPN in QubesOS.
I have not tested it with the Fedora templates, but both the bridge and VPN works just fine with Debian.
Split tunneling works with sys-net-vpn or sys-net for different applications.
Cool?