Allowing me to pay would be yet another right direction for Telegram, but i just can't move past the audit (and to a lesser extent, FOSS)
I'd go as far as saying it's by far the best. I have used the word unnerving to describe the fact it's so beyond the rest of the competition, yet free and without a viable business model.
I use Signal with friends and Matrix with other nerds.
Given the open-source nature of tg, I would imagine someone would put together a Messenger client based on tg source code. I guess that would solve many problems with FB's own messenger.
This has been bothering me as well but seems to be solved now.
All current messaging features will be free but new team features etc will be paid IIRC and AFAIK.
Well, we'll have to see how that pans out...
Do you refer to the server-stack? How would that being Open Source help you when you cannot ever verify what a server actually runs?
Primarily yes, and while true - that's an extreme and a bit pointless, imo. I can say the same about Linux, which i'm on now, as i've never verified what i'm _actually_ running. Or Matrix servers i'm connected to, and what they're _actually_ running.
Yet the idea remains that something some people have analyzed is better than something "no one" (outside Telegram lol) has analyzed.
Your question feels as if we may as well all be running entirely closed source. I'm unsure why the benefits and/or supporting arguments for FOSS need to be stated here. Is there a degree of your question i'm missing?
edit: Oh and, of course, the FOSS nature is even more important in the case where the majority of Telegram communication seems to lack security. If this was a zero knowledge platform i'd be far less concerned about their serverside implementation.
I understand the confusion. I was talking about the current situation where you cannot run your own server or even choose amongst servers.
It is pointless if you can view the source-code of that server, but have no way to check if the one server that you can ever use, runs that code at all.
It makes sense in a federated or decentralised setup, where you can run your own servers, choose instances, or even build your own version of the client with other backend-urls baked in. For Telegram all it offers is validation that the code is good, or not good. Without any power to do anything about that.
That's literally impossible with any closed-source server-side software. You can't even inspect the binary as you can with a client.
And here's the server implementation you can also run yourself: https://github.com/signalapp/Signal-Server
Telegram has also been exploited multiple times, while signal has held up against a US federal subpoena.
https://www.theverge.com/2016/10/4/13161026/signal-subpoena-...
[1] https://calyxos.gitlab.io/calyx-fdroid-repo/fdroid/repo?fing...
Just that Signal is not on F-Droid. Which has a bit of a history and some drama. It was there, log ago. Still textsecure back then.
Sure, you can build it yourself. And from there, with some hoops to jump, install it on your phone over adb (it's what I did). But that is far from fdroid or some other app store. E.g. there are no updates, which is a crucial feature in a security-critical app.
Source?
Where possible, I prefer to type on my proper desktop keyboard, because I am a stubborn old man and you kids with your phones, and Telegram is just so nice in that context. It uses no resources, the UI largely gets out of the way, and message sync and delivery is 99.9% reliable in my experience. (Looking at you, Messages app.)
I do worry on some level that, by not paying, I am the product in some sense. And I wish they had a full security audit. But, I have so many other potentially insecure ways for people to get my info, and really I’m too boring to stalk anyways, so I just can’t make myself too bothered.