!wzHrsErnsyaqbpFiRQ:matrix.org

Ruma

653 Members
Discussion of Ruma, a Matrix homeserver, client, and supporting libraries written in the Rust programming language. https://www.ruma.io/166 Servers

Load older messages


Timestamp Message
16 Sep 2019
09:41:00@jplatte:matrix.orgjplatteSorry, key signing is probably the wrong term
09:43:16@jplatte:matrix.orgjplatte But I don't want Riot nagging me to trust devices I logged in with. Look at my device list, you can imagine most of those aren't actually permanent logins on a personal device.
09:44:01@jplatte:matrix.orgjplatte

Ah so this binary user data doesn't follow a spec? Would be annoying to have every client use a different system for that if it came to such a feature

I think it's arbitrary JSON, not binary data

11:15:34@andrewm:amorgan.xyzanoa
In reply to @jplatte:matrix.org

Why do you consider the E2E design unsuitable?

My main gripe is that all my "devices" are equal. I can't tell matrix on login that the device I'm using right now should not receive private messages and should not sync keys with other devices. Also I find it not so nice to have all private messages kept on the server, even if encrypted.

The upcoming work on cross-signing helps deal with attackers logging in to your account and preventing them from having keys synced to them. This sounds like a similar usecase where you just want to use throw-away devices, but not entrusted them with e2e keys. Hopefully the UX will update to match what you're after.
11:16:52* @jplatte:matrix.orgjplatte shrugs
11:17:52@jplatte:matrix.orgjplatteHonestly, I don't intend to try getting my 1-to-1 conversations onto Matrix anytime soon even if e2e UX is improved
11:20:55@jplatte:matrix.orgjplatteThere would have to be something wrong with Signal first, plus app performance would need to be on par (which I guess this project kind of is about, but of course we're still very far away from a full client or server written mostly in Rust)
12:59:35@sphinxc0re:das-labor.orgsphinxc0re changed their profile picture.
13:47:52@paulvt:matrix.orgpaulvtSignal has centralized servers right? I'll stick with XMPP for now
13:52:22@jplatte:matrix.orgjplatteYeah, it does. Which IMO is really an advantage UX-wise, in todays world. I would really like federated services to be more mainstream, but currently they tend to just confuse non-technical people.
22:09:39@jplatte:matrix.orgjplatte Jimmy: Working on ruma-events I noticed that you always implement Serialize manually. Why is that?
22:11:49@techranger:matrix.orgtechranger joined the room.
23:27:19@asonix:asonix.dogAode (Lion) He/They changed their profile picture.
17 Sep 2019
04:33:22@jimmycuadra:matrix.orgJimmy
In reply to @jplatte:matrix.org
Jimmy: Working on ruma-events I noticed that you always implement Serialize manually. Why is that?
IIRC, manual implementations were to make sure that field counts for structs are correct, since not all of the fields come from the definition at the proc macro call site. if it's possible to get the same result using derive, that'd be fine. maybe I just overlooked something.
06:19:19@freenode_Fanaen:matrix.org@freenode_Fanaen:matrix.org left the room.
06:19:21@freenode_vinipsmaker_:matrix.org@freenode_vinipsmaker_:matrix.org left the room.
06:19:22@freenode_craigp:matrix.org@freenode_craigp:matrix.org left the room.
06:19:26@freenode_spossiba_:matrix.org@freenode_spossiba_:matrix.org left the room.
06:19:29@freenode_raynold:matrix.org@freenode_raynold:matrix.org left the room.
11:20:08@freenode_tollyx:matrix.org@freenode_tollyx:matrix.org joined the room.
17:08:10@freenode_Bonstra:matrix.orgBonstra left the room.
17:11:29@freenode_Bonstra:matrix.orgBonstra joined the room.
18 Sep 2019
01:20:44@finaltraveler:matrix.orgfinaltraveler joined the room.
12:35:32@mrwimpy:matrix.org@mrwimpy:matrix.org left the room.
19:10:02@kltx:matrix.org@kltx:matrix.org left the room.
19 Sep 2019
09:34:27@hubert:hirtzfr.euHubert changed their profile picture.
14:55:22@freenode_tollyx:matrix.org@freenode_tollyx:matrix.org left the room.
17:07:04@jplatte:matrix.orgjplatte Jimmy: Another question about the event API: do we want / need the fallback to a raw content struct in any context where there's not an event struct around the content that also falls back?
20:58:47@jimmycuadra:matrix.orgJimmy
In reply to @jplatte:matrix.org
Jimmy: Another question about the event API: do we want / need the fallback to a raw content struct in any context where there's not an event struct around the content that also falls back?
it might be necessary for cases like stripped state events that use the content structs with different wrappers
21:03:36@jplatte:matrix.orgjplatteInteresting, okay.

There are no newer messages yet.


Back to Room List