WebRTC is a protocol, you can use it in both web apps and mobile apps.

If I remember well it was used as a trick from Facebook to leak personal data using the SDP.

“3.3 Use Limitation” section of their licence file is not compatible with open source definition.

That’s vibe coding.

What about Bitchat?

We should support our fediverse admins and instances 💪 Support by sending money (for people who can), moderating content, submitting issues or helping the team and project ✌️

So abandon open source and move to “post open source” or ethical source might be a (sad) solution.

It is complicated. He says true. And maybe your need to use GrapheneOS is relevant. If you have a smartphone without cellular connection, for a daily usage, FMPOV it is a non-sense in case of emergency. It is a risk you will have to take, I can’t disagree your dad. And what about your solution but with a SIM card with very few data and SMS available, through a SIM card you can keep aside your phone and insert when needed?

Anticipate technical debt and follow what Google recommends. In few words, use Kotlin and Compose.

However you should really have a look on Google guidelines. In more worlds:

• by default Kotlin and Compose
• if some logic to share between other projects in other environments: Kotlin Multi Platform (KMP)
• if shared UI: Flutter (but Google reduced Flutter teams and KMP is being better and better, so we can suppose Flutter will join the Google Graveyard

I do not know if the solutions I listed below are open source ; however as an open source contributor I am used to work with some tools depending to choice of the projects:

• Crowdin
• Weblate
• Loco

About credits, I don’t think these tools exposes in some automated way the contributors identities. However, nothing prevents you to use these web UI tools to find who contributed and list people for example in your CONTRIBUTORS files. Another way could be to edit the automated commits these tools submit to your Git repos by adding credits to the translators (with for example Co-authored-by field).

Did you have a look on Cake Wallet app? Open source under MIT license and available on F-Droid.

https://github.com/cake-tech/cake_wallet

Yep, it seems it is, but it can manage KDBX files. Just wanted to share 😄

Edit: sorry, didn’t see this thread is in Android community, my comment is not relevant for this platform.

You can use also for example Strongbox (https://github.com/strongbox-password-safe)

Edit: sorry, didn’t see this thread is in Android community, my comment is not relevant for this platform. For Android I am used to Keepass2Android (https://github.com/PhilippC/keepass2android). Simple, still maintained, under libre licence GPL 3.0.

It is always the same issues in fact. You should consider your threat model before all. Then, consider the Signal app, then your iPhone supposed to be updated, trusted, with ADP enabled, biometric lock with erasure after 10 failures, etc. Then consider your ISP, then your country. Etc, etc. You should also compare the contexts. Is an iPhone “better” than a low or middle ranges Android-powered smartphones? For sure, yes. Is it better than high-range expansive smartphones with Android ? Or Pixel ones? Not that sure. And compared to GrapheneOS or /e/? Pretty sure not that much. You can also compare messaging solutions. Is Signal better than WhatApp? Of course yes. But what about XMPP and Matrix for example?

And what are your use cases? Remember your threat model. If you are an activist, a journalist or a whistleblower your needs may be different than a “commons citizen worried about its privacy.

In few words, the only pain point I see is the fact than iOS is proprietary and runs non libre source code and Apple devices than APN. But Android devices are not so much different. It does not mean the solution is not private or efficient, if we succeed in defining a definition of “private or efficient”.

In a nutshell, it could be considered as good. But not perfect.

Any ideas for E2E encrypted storage alternatives?

Not sure of that, maybe we need some case law or update on existing copyleft licenses. Source code generated with GenAI tool, even if their model have been trained with corpora of copyleft sources, are not (yet) considered as derivative works. What a pitty.

Could be interesting. Non-free and current GenAI tools violate copyright, we may consider some evolutions of copyfarleft licenses to forbid such use of source code in these types of tools.

And nice comment spotted there: https://fosstodon.org/@geraldew/113849843708286036

Just wanted to share for the common knowledge and the debate as I already saw here some “post open source” and content about rubbish licenses like SSPL or BSL 😉

Be sure also the issues you have in your project have the suitable labels to help future contributors to pick easily some of them, i.e. labels like “help wanted” or “good first issue”.

You can also refer to best practices listed and explained for example in Advent of Open Source so as to have a nice and user-friendly repo: https://adventofopensource.com/

Have a look on Organic Maps (https://organicmaps.app) or OSMAnd for example (https://osmand.net/).