[messaging] Panoramix decryption mixnet messaging spec and design documents

[dawuud]

Hi Michael,

> Ah, thanks for explaining. I thought the goal was to build a
> general-purpose mixnet infrastructure on which people could build
> as-yet-unknown applications. Hence all the questions about what options
> are available to applications and what interfaces are exposed by
> different layers.

You will have to write a new client and modify the Provider a bit to
get it working with your application. The Provider isn't written
yet... (nor is the mix) but I'm hoping Yawning will get to work on it
soon. As for the client component... it's fairly easy to write one
reusing much of our existing code. For instance our "core" repo has
our wire protocol library and a Sphinx library. The "client" repo has
our crypto/block library for end to end messages using noise_x. The
stuff that you don't want such as POP3 and SMTP can easily be omitted
from whatever client you end up writing.

> But even if the goal is just to support specific applications, I wonder
> if it would be useful for clients (by which I mean mixnet clients, not
> MUAs) to have some control over decoy traffic, because different clients
> will have different constraints. For example, a desktop client may
> prefer different power/bandwidth/anonymity tradeoffs to a mobile client.

Well, it depends what you means by "client"... it's gotten rather
blurred with how we've designed things. I would say that the mixnet
client *does* have control over the decoy traffic it emits; how would
it not have control? On the other hand I don't think users should be
burdened with these details. (then again i don't really want to
discuss user interfaces but if we were to discuss this then I would be
citing papers written by Ka Ping Yee and pointing out design
principles such as the principle of least resistance etc.)

Let me know if you have any questions about the specifications after you read them.


Cheers!
David

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: not available
URL: <http://moderncrypto.org/mail-archive/messaging/attachments/20170921/caa3111c/attachment.sig>