indy-did-method
indy-did-method copied to clipboard
Resolvable domain name for namespace
Currently, DID Indy Namespace
is defined as "a string that identifies the name of the primary Indy ledger".
I would like to propose that the "string identifier" to be a resolvable domain name where the genesis can be self-discoverable in distributed manner.
For instance:
- from
sovrin
tosovrin.org
- from
idunion
toidunion.org
Since a resolvable domain name is now part of DIDs, the organization in control of the domain is also in control of their genesis file which can be discoverable by using well-known URI, such as
- http://
<nnamespace>
/.well-known/hyperledger-indy-ledger/<namespace identifier>
/metadata.json - http://
<nnamespace>
/.well-known/indy-ledger/<namespace identifier>
/metadata.json
I am proposing a metadata.json
file as it can also be used to add metadata information about the ledger (e.g.: production vs non-production).
Initially, I think that providing the github repository is still a good way for wallets to start using indy method while ledgers have time to catch up and provide the necessary endpoints.
Let's discuss this at the Indy Contributors call this coming Tuesday.
Agenda: https://wiki.hyperledger.org/display/indy/2022-04-26+Indy+Contributors+Call
I believe some examples would be:
- Sovrin MainNet:
did:indy:sovrin.org:3pBgqwcAm5SXqgPop8J8gH
- Sovrin StagintNet:
did:indy:sovrin.org:staging:3pBgqwcAm5SXqgPop8J8gH
We discussed this a bit and i like the idea to use a domain name, e.g.
- did:indy:sovrin.org:abc
- did:indy:idunion.org:abc We could still use a github repository governed by hyperledger but this would give us the freedom to use both options. Does anyone have a good reason not to do this?
Discussed at the Indy Contributors call 2022.08.30. Agreement:
- Update did-indy-method specification to include this.
- Make it optional – the challenge for some will be in retaining a long lasting domain name
- How to keep the two versions (domain hosted, one in did-indy-networks repo) of the genesis file in sync?
- Idea – GHA that compares the two files and if they differ, creates a Pull Request to update the one in GitHub repo to match the one on the domain name.
- Pull request must be reviewed and merged by the editors according to the governance – ensures there is a human step in the process
@swcurran , where are we at with this suggestion? I haven't seem any updates to the documentation