3 ms·
For some of us, "the organization controlling the central infrastructure" is _way_ more trustworthy to not start "acting poorly" than any of the alternatives.
by bigiain 6y ago
For some of us, "the organization controlling the central infrastructure" is _way_ more trustworthy to not start "acting poorly" than any of the alternatives.
For me, Apple comes a close second maybe, but lack of interoperability between iMessage and Android makes it a non starter amongst my friends/family. Even assuming some self-hosted version of an E2EE messaging service exists and I could convince enough of my family/friends to use it, I then become "the organization controlling the central infrastructure" who risks "acting poorly" due to incompetence or lack of resources to keep that self hoisted infrastructure running and secured.
Signal is not perfect. I don't agree with all of Moxie's choices (I'd strongly prefer it not to need to be linked to a real world phone number) and I strongly disagree with some of his choices (I get angry every time a "$name (someone in my contact list) has just joined Signal" notification arrives.
But it's better than the alternatives for me. And for enough of my friends/family that it's my most commonly used comms channel outside work.
- cheph 6y agoTo me the end result of this shifting landscape is something which has the attributes of Matrix. Matrix may have issues, but it's architecture and implementation is very resistant to bad actors. You can fork the server code and the server instances of matrix servers and have it work with existing servers and clients. Without this capability it is a matter of time until bad actors kill it and everyone moves to the next thing. The problem is that "trust" is not enough. You trusting Signal operators more than Whatsapp operators does not fix the problem that we cannot run these services on trust.