Re: State of the decentralized groupchat

[TheJackiMonster]

Hey t3ss,

On Wed, 2020-09-02 at 10:02 +0200, t3sserakt wrote:
> 
> Definitely! Then an application can handle different egos, too. This
> is
> a nice feature, if you like to use the cadet-gtk/groupchat to
> communicate with egos of other persons and you do not like them to
> know
> your egos are one and the same person. The "only" missing piece then
> is
> to have routing on top of cadet which is not using peer ids for
> routing,
> because in case of routing based on peer ids it is known to other
> peers
> that messages for two different egos are send to the same peer. 

the routing works currently as all new messages are forwarded to all
peers you are connected to unregarding of their peer id. The only one
who can match peer id and your ego key is the one peer you have
connected to join the chat room.

The other possibility to reveal your peer id to others is a specific
message you can send which will only be sent to others if you act as a
host in the chat room and become part of the ring topology (that
feature requires others to know your peer id).

But you could still connect to two different peers in the same room
with two different egos. Even if one of them would act as a host others
only one other peer in the room could match both egos to relate to the
same peer.

It will also be possible to use the same ego key with two different
peers and share the identity. This would allow using a chat client on
different devices even in parallel.

Greetings
Jacki

signature.asc
Description: This is a digitally signed message part