[kwlug-disc] Videoconference testing -- please help

[Hubert Chathi]

On Mon, 13 Apr 2020 14:47:46 -0400, Paul Nijjar via kwlug-disc <kwlug-disc at kwlug.org> said:

> On Mon, Apr 13, 2020 at 10:46:56AM -0400, Hubert Chathi wrote:
>> On Tue, 31 Mar 2020 22:23:45 -0400, Hubert Chathi <hubert at uhoreg.ca>
>> said:
>> 
>> > Pretty much no video conferencing system that supports a large
>> number > of participants will have end-to-end encryption.  The
>> problem with > Zoom here is that they're claiming that they do.
>> 
>> Jitsi is working on end-to-end encryption, and has a usable
>> proof-of-concept: https://jitsi.org/news/e2ee/

> How does Matrix fit into this? I thought its own deal was to be an
> everything to everything communications platform with privacy as a
> focus?

Matrix itself just has 1-1 calling at the moment.  For conference
calling, Riot uses an embedded Jitsi.  When Jitsi has fully-working
e2ee, we (Riot) will likely try to integrate it into Matrix's e2ee
system, so that it uses Matrix to distribute the keys, and benefits from
the key verification work that we've been doing [1].

Ideally, everything would happen in Matrix itself, but that conference
calls likely won't happen for a while, and we'll continue to use Jitsi
for some time.

On Jitsi's side, they're looking into using Matrix's implementation of
Signal's double-ratchet system for key distribution, though I don't know
exactly how that's going to work.

[1] You can watch a demo of our work-in-progress key verification at
https://www.youtube.com/watch?v=APVp-20ATLk&t=6120s which comes from a
recent open tech meetup that we hosted.  (Watch for a release coming
very soon.)  Also, if you want to see how the Jitsi team have been
handling increased traffic over the past few weeks, you can watch an
earlier talk from that same meetup:
https://www.youtube.com/watch?v=APVp-20ATLk&t=850s