Zoom meetings aren’t actually end-to-end encrypted, despite misleading marketing on their website, in their security white paper, and in the user interface in their app https://theintercept.com/2020/03/31/zoom-meeting-encryption/ by @yaelwrites and myself
@micahflee @yaelwrites zoom works with 10+ simultanous video streams because it uses Multipoint Control Unit (MCU) so by definition it needs to access all video streams (to remix and efficiently redistribute them); jitsi uses SFU so it is E2E encrypted, but as more and more participants join the bandwidth is all taken https://medium.com/linagora-engineering/scalability-in-video-conferencing-part-1-276f52b4acac
@charlag @micahflee @yaelwrites thanks for the correction; according to the devs jitsi meet is not E2E encrypted except for 1:1 videconfs with p2p mode on https://github.com/jitsi/jitsi-meet/wiki/Jitsi-Meet-Encryption
@kekcoin @micahflee @yaelwrites youre right, I had already amened my toot here: https://mastodon.social/@paolog/103918744933790563
@linos @kaffeeringe @micahflee @yaelwrites That shouldn’t be an issue. I mean you can easily download stuff with an iPhone at Gigabit speeds, which also has to be decrypted if it comes via HTTPS, so it’s not a problem for any somewhat modern computer.
FaceTime allows big group calls and it is E2E encrypted, so there must be a way.
@micahflee @kaffeeringe @melgu @linos @yaelwrites I guess I would see the "host" as another (ad-hoc) "server" in that setup (being the server for the videoconference itself, whereas what you call the server simply hosts the software and potentially serves as a STUN server).
I certainly wouldn't consider it P2P, though I also see how it has some advantages over a traditional model where there is only one server (mostly privacy-related).
Server run by the main developers of the project It is not focused on any particular niche interest - everyone is welcome as long as you follow our code of conduct!