| 1 Dec 2024 |
 Nico | Well, in that case you can't use it on a phone | 05:02:21 |
 Solodric | In reply to @ity:itycodes.org
What we know agrees with what Nico said about Tor traffic correlation This is the commonly-available understanding. The problem is that the Tor project only mentioned their new timing algos in a single patch/update press release in 2016, and has barely talked about it since. I had to run down some actual Tor devs to ask about it :/ | 05:02:29 |
 Tranquil Ity | Welp | 05:02:41 |
| Nico | You can't have all three of timing safety, low latency and low data usage | 05:02:51 |
| Nico | iirc | 05:02:53 |
| Solodric | In reply to @deepbluev7:neko.dev
You can't have all three of timing safety, low latency and low data usage This is true! I hadn't even thought about data limits, but you're absolutely right.
Although not everyone's phone has a data limit problem.
| 05:03:17 |
| Tranquil Ity | In reply to @deepbluev7:neko.dev
You can't have all three of timing safety, low latency and low data usage Yea | 05:03:27 |
| Nico | No, but a battery | 05:03:31 |
| Solodric | Hah! that's true too. | 05:03:40 |
| Solodric | You can tell I barely use phones. | 05:03:44 |
| Solodric | I2P would definitely put a strain on your data and battery. | 05:03:58 |
| Tranquil Ity | Ours' battery is dead | 05:04:00 |
| Tranquil Ity | Wish we could change batteries easily | 05:04:12 |
| Nico | Matrix isn't the end all be all. It is somewhere on the secure to convenient scale | 05:04:20 |
| Solodric | The new Purism phones might actually tempt me to use a phone if they become affordable. | 05:04:23 |
| Tranquil Ity | In reply to @deepbluev7:neko.dev
Matrix isn't the end all be all. It is somewhere on the secure to convenient scale Yea ^ | 05:04:35 |
| Nico | There are examples going further in either direction (or both) | 05:04:40 |
| Tranquil Ity | In reply to @solodric:matrix.org
The new Purism phones might actually tempt me to use a phone if they become affordable. We are NEVER touching Purism | 05:04:42 |
| Solodric | In reply to @deepbluev7:neko.dev
Matrix isn't the end all be all. It is somewhere on the secure to convenient scale Yeah. That's why I'm looking at it.
If you want totally bulletproof anonymity, you use P2P over I2P.
| 05:04:44 |
| Tranquil Ity | A friend of ours used IRC P2P over Tor lol | 05:05:01 |
| Tranquil Ity | Made us design a protocol specifically for that | 05:05:10 |
| Tranquil Ity | * Made us design a protocol specifically for that usecase | 05:05:11 |
| Nico | Anyway, gotta go, good luck you two | 05:05:16 |
| Solodric | that's wild. | 05:05:17 |
| Tranquil Ity | https://itycodes.org/directchat-01.html | 05:05:18 |
| Solodric | In reply to @deepbluev7:neko.dev
Anyway, gotta go, good luck you two Thanks for the help! | 05:05:25 |
| 2 Dec 2024 |
| Solodric | Would it be possible to use a protocol similar to Oblivious DNS Over HTTPS to reduce the amount of unencrypted metadata on Matrix? While I don't think such a method could be bulletproof, you could potentially distribute trust between multiple parties in a Tor-esque fashion. | 04:55:22 |
| Solodric | At the bare minimum, it seems like you could use this methodology to minimize what a given homeserver admin can see. | 04:56:28 |
| Solodric | But I might just be foolish for suggesting it. I haven't given careful consideration to which specific metadata would be concealed or how, it's just that this is a known way to try to conceal metadata that I realized could work for an entirely different project (Fediverse stuff) | 04:57:09 |
 richvdh | the principal problem for metadata on Matrix is that all servers have to know the full list of members in a room, so that they can correctly fan out new messages to all the other servers in the room | 10:23:26 |
