Open Discussion on Email, Messaging, and SSI/DID
Open Discussion on Email, Messaging, & SSI/DID
Wednesday 17A
Convener: Ryo Kajiwara
Notes-taker(s): Ryo Kajiwara
Tags for the session - technology discussed/ideas considered:
Discussion notes, key understandings, outstanding questions, observations, and, if appropriate to this discussion: action items, next steps:
Slides: http://speakerdeck.com/sylph01/did
Comments:
There is a similar solution that is PGP, but nobody has figured out how to use it
- There is a DIDComm app that does something close along these lines, but through the use of this app, there were no mentions of “long numbers” by the users of this app
- We need to have a streamlined user experience, otherwise it would lead to “corporate capture” (as mentioned in the “Fundamental Problems of Decentralized Systems” session)
- Email might be already under corporate capture; we need to please the great email providers to make them accept a new extension/alternative to email
There are two different problems/use cases mentioned:
- Encryption and privacy -> Applies more to person-to-person use case
- Veracity of the sender -> Applies more to unsolicited communication
How much cannot be done with existing solutions?
- Need to dig down on this, but PGP and S/MIME partially solve the issue
How could we layer this using the existing infrastructure?
- Using a header extension to carry VC/DID associated information in the current mail headers can be a transitional solution moving forward
In the personal messaging space, encryption is more important, but in the case of advertising and notification, it gets less important The use case where you know each other (-> use case with “pre-established trust”) is easy; where we use email today is to send email to a completely unknown person. Can we do that with SSI/DIDs? Are these identifiers stable enough or understandable enough?
- We might need a (partially) centralized directory to find “addresses” of someone
Using VCs/DIDs to identify the sender can be thought of as replacing “pre-established trust” with a trust mechanism that can be globally used
