-
nicoco
> any idea why https://koishi.pain.agency/matrix-proxy/0bc9b146-bf41-4bd1-a26b-996d2d4df287/image.png wont embed in an xmpp client? if i curl it to a file it seems correct but gajim will just dissapear the download and dino blanks out the message 503 bad gateway for me ↺
-
franco
Do you know how the decrypted payload looks like in OMEMO? Is it an XML stanza? Plaintext? Something else?
-
lovetox
https://xmpp.org/extensions/xep-0384.html#example-6
-
lovetox
But beware there are older versions of the spec where just the message string was encrypted and not xml
-
moparisthebest
and by "older versions" you mean the one gajim and everything else supports right now, vs the link which maybe 1 client supports (:
-
franco
That's very useful insight, thanks✎ -
franco
Those are very useful insights. Thank you both ✏
-
wgreenhouse
in the most commonly supported o.3 version, only the <body> tag of the overall xml message stanzs is encrypted
-
wgreenhouse
which is why things like reactions currently pose a leak, if allowed by your client during an OMEMO session
-
wgreenhouse
*stanza
-
wgreenhouse
*0.3
-
AZERTY keyboard
I will look at my logs to see
-
AZERTY keyboard
For omemo it was in a <payload> which is inside <encrypted>. While reaction was an unencrypted <reaction> with a stanza.
-
franco
I'm trying to figure a way to keep E2EE through a bridge if the other end also implements Signal Protocol for E2EE (for ex., Matrix). Maybe if clients supported converting the payloads from/to XML/JSON, after decryption 🤔
-
franco
I've looked at mautrix-whatsapp (which bridges Matrix and WhatsApp), but they don't preserve the E2EE (the bridge is a MitM).
-
stratself
I really don't think any bridge could do E2EE unless protocols pre-agree with what to send
-
stratself
which is what IETF MIMI is supposed to solve
🤝 1 -
singpolyma
it would have been possible with oldmemo but yeah, not realistic in general unless both sides speak the same "inner format"
-
franco
Is anyone here attending the MIMI WG meetings? I'd like to