Re: did:twit method

I, for one, welcome the re-emergence of Keybase-like elements in a decentralized way using DIDs and VCs! Kudos for also bringing the code along with the spec. Will we see did:hn or did:w3 next? :)

If we begin to accumulate varieties of powerful centralized non-colluding non-custodial DID methods does that add decentralization to the system along a different axis? We could have schemes where an attacker must compromise 2-of-3 2FA Internet accounts to attain DID Controller privileges. What could be the transition/interoperability story with fully self-custodied DIDs such as btcr?

Shamelessly plugging the DID decentralization rubric some folks on this list sunk good work into:
https://docs.google.com/document/d/1rYdWiwawWmLOWtHRvT0GzYcdewW_OS9M2mAkENLFdtY/edit?pli=1#heading=h.4p260dq0jbu

On Fri, Jul 31, 2020, at 7:09 PM, Gabe Cohen wrote:
> Hi everyone,
> 
> I've created a fun side project introducing Twitter-based DIDs. 
> You can find the spec here, inspired by both the did:key and did:github methods: https://github.com/did-twit/did-twit/blob/master/spec/index.md
> I have a cli where you can build did:twit identifiers, author messages, and verify other's did:twit messages here: https://github.com/did-twit/did-twit-cli
> 
> Imagine if everyone affected by the Twitter hack a few weeks ago had been using verifiable tweets? Some ideas I have for where to take the project include storing the signatures on IPFS to reduce the noise on Twitter, or building a verifiable Tweet client.. Reach out if you're interested in contributing.
> 
> Gabe