Re: Can we keep did-method as a variable in the spec?

Hello Vishwas,

The DID Working Group has discussed this topic in the past, about 
potentially having "DID method types" or "abstract DID methods" that can 
be used as a basis for multiple DID methods, e.g. see discussions here:
https://github.com/w3c/did-core/issues/152
https://github.com/w3c/did-core/issues/478

The conclusion was that this is probably not a good idea, and that a 
single DID method (such as "hypersign") with sub-namespaces should be 
used. In your syntax you already using something called a 
"network-namespace", and I think the "zones" should just be another 
namespace before that, e.g.:

did = "did" ":" "hypersign" ":" zone ":" network-namespace ":" unique-id

Then you don't need separate DID methods in the W3C DID method registry. 
Do you see a problem with that?

I think the "test" should be whether or not there is a substantial 
difference in how the DID operations create/read/update/deactivate work 
technically. If those operations work essentially the same way across 
different zones and networks, only with different parameters or 
configuration, then you don't really have separate DID methods.

Other DID methods such as did:indy or did:ethr are doing something 
similar, they use sub-namespaces for identifying different networks, or 
zones, etc.

Also see the green note box at the end of this section:
https://www.w3.org/TR/did-core/#method-syntax

Markus

On 17.02.22 10:35, Vishwas Anand wrote:
> Hi all,
> We are building Hypersign using Cosmos SDK whose purpose it to manage 
> digital identities in Cosmos ecosystem and any other zone should be 
> able to resolve DID via IBC bringing interoperability within zones. We 
> are going through W3C spec and are in the process for defining our/did 
> method and registering it in the W3C did registry/. We have some 
> confusion regarding this which we would like to clarify.
> The Cosmos ecosystem consists of application specific blockchains, 
> called zones. These blockchains/zones are for specific purpose, built 
> for one use case. We can visualize each smart on Ethereum being a 
> separate blockchain having its own governance. Any one can build a 
> zone using Cosmos SDK which provides base framework for building a 
> chain and then one can implement their use case on top of this SDK. 
> The Cosmos Zones also have way to interact with each other using a 
> tech called IBC – Inter Blockchain Communication.
> At first glance it looks like Hypersign going to be a centralized 
> blockchain which will manage DIDs for all zones in the cosmos 
> ecosystem. But the way we are visualizing is, once we have built the 
> SSI module properly for our zone, we will request Cosmos SDK to merge 
> this module as a native module. This way any zone can enable SSI  
> module in their network./I believe that in future there will be 
> multiple registstires instead of one centralized place for DIDs./A 
> issuers may exists on zone 1 whereas a verifier may sits on zone 2 and 
> vice verse.  DIDs issued on one zone can be resolved on the other to 
> achieve interoperability. We feel this would be great step towards the 
> adoption of SSI  technology.
> The problem is, if a zone need to become a registry  then they also 
> need to properly implement the SSI principles by going through W3C 
> spec and then they need to register their DID-method in the registry. 
> We know that is very tedious task and beside this it will distract 
> them  from building their own application specific use case / business 
> requirements.
> I was wondering can it be possible to have template or a variable 
> for/method-name/ something like this?
> /did                  = "did" ":"  method-name  ":" network-namespace 
> ":"  unique-id/
> /method-name              = zones/
> /zones                             = “zone1” / “zone 2” / “zone3” / …. 
> / “zone n”/
> /network-namespace   = "main" / "test"/
> /unique-id                       = 32*255id-char/
> /idchar                             = ALPHA / DIGIT/
> Like, we can register a did-method with list of known zones. In future 
> if a new zone is added to the cosmos ecosystem , then they can just 
> raise the PR for the same spec to add their zone names in the list 
> of*zones*. Can it be possible to do so ? and will it be feasible from 
> the security point of view? Or does it violates SSI principles?
> Sorry, if this idea is too naïve. But wanted to clarify it. Looking 
> forward to hear your thoughts. TIA.
> --
> Vishwas Anand Bhushan
> Co-Founder @Hypermine Technologies <https://hypermine.in/>| R&D Engineer
> /BuildingHypersign <https://hypersign.id/>/–/An////interoperable 
> decentralized identity blockchain network/