Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

RFC 0211: routing shouldn't be exclusively key-focused #274

Closed
dhh1128 opened this issue Oct 29, 2019 · 9 comments
Closed

RFC 0211: routing shouldn't be exclusively key-focused #274

dhh1128 opened this issue Oct 29, 2019 · 9 comments

Comments

@dhh1128
Copy link
Contributor

dhh1128 commented Oct 29, 2019

The way 0211 is currently written, all routing is specified in terms of keys. I think this is too narrow. It may be reasonable to do intra-domain routing from Alice's cloud agent to her edge agents using keys, but at least inter-domain routing from Alice's agency to her cloud agent will likely use DIDs.

@devin-fisher @TelegramSam

@kdenhartog
Copy link
Contributor

Would it be a DID or would it be a DID-url? (DID with key reference) Is this something we want to specify the difference between?

@dhh1128
Copy link
Contributor Author

dhh1128 commented Nov 4, 2019

It would be a DID. I am saying that when Alice wants to send to Bob, she routes to Bob's DID. She may encrypt for Bob's edge agent(s), but encryption and routing are not the same (though they are highly related).

I'm going to write some more about this logical difference in the new RFC about routing.

@llorllale
Copy link
Contributor

@dhh1128 @devin-fisher @TelegramSam @kdenhartog was this topic discussed elsewhere? We share Daniel's concerns as well.

@llorllale
Copy link
Contributor

The to field of the Forward message accepts DIDs with and without key references. The route-coordination RFC seems to be misaligned with this fact.

Any updates on this?

@TelegramSam
Copy link
Contributor

We went with keys initially, as allowing either plain DIDs or DID URIs (that point to keys) adds complexity. Now might be the right time to expand to something more expressive. We'll need to consider the solutions to the edge cases that this brings up.

@TelegramSam
Copy link
Contributor

One complexity of using DIDs is that the list of keys in routing keys is used for encrypting messages. If we use DIDs, which key do we use during encrypting a message for that step? This is only an issue with Mediators (known to sender) and not an issue with relays (not known to sender).

@TelegramSam
Copy link
Contributor

Important question: Should we leave it keys for now, and fix this in the DIF DIDComm WG work, or do we need to fix it faster for the Aries community? @dhh1128 @llorllale

@dhh1128
Copy link
Contributor Author

dhh1128 commented Jan 30, 2020

I'd say let's work on it in the DIDComm WG, and if a specific need arises to do something sooner, we can also work on it in Aries. But let's not work on it in Aries until we know we need to.

@TelegramSam
Copy link
Contributor

Continue issue here: decentralized-identity/didcomm-messaging#26

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

4 participants