Minutes interim-2024-mimi-01: Wed 16:00
minutes-interim-2024-mimi-01-202404101600-00
Meeting Minutes | More Instant Messaging Interoperability (mimi) WG | |
---|---|---|
Date and time | 2024-04-10 16:00 | |
Title | Minutes interim-2024-mimi-01: Wed 16:00 | |
State | Active | |
Other versions | markdown | |
Last updated | 2024-04-12 |
mimi interim - 10 April 2024
Architecture Document
No open tickets
Protocol Document
-
#72: Add consent discussion
- Richard: Merge? Yes
-
#73: Abuse reporting
- Tim: Can abusive content be forged? (Rohan: maybe?)
- Travis: Should reports be sent to hub or provider?
- Brendan / Raphael: Franking needed.
- Travis: Exactly where report should be routed depends on the
type of abuse. - More review needed before merging.
Metadata Minimization
Brendan: What do you mean when you say you don't provide anonymity?
Konrad: Full handshake encryption would provide stronger privacy
properties. This has an easy link between activity and real identity,
hence pseudonym.
Rohan: If you don't have connections, could you use a non-pseudononymous
credential? Join with pseudonym and delay revealing connection to real
identity?
Konrad: Sure, those are options
Rohan: I would forgoe using the target user's serivce provider's
directory of KeyPackages. I would only use KPs that were provided
directly.
Brendan: Would encourage the TreeWrap idea to be applied directly to the
ratchet tree.
Protocol Document
- #74: Identifiers primitive
- Richard: How does this relate to discovery?
- Rohan: Provider discovery is about taking an identifier, finding
the provider. This finds internal identifier (uuid) from
external identifier (email/phone/etc) - Tim: Why do I need to know the type of an identifier?
- Rohan: Context improves search quality
- Tim: GET request, but has a body
- Rohan: Will fix
TLS vs CBOR
Rohan: Minimal efficiency difference
Tim: Will continue discussion on mailing list