Clarity on Section 2.1

[nadalin]

"These new identifiers are designed to enable the controller of a DID to prove control over it and to be implemented independently of any centralized registry, identity provider, or certificate authority"

I assume that a DID can be used by a centralized registry, identity provider, or certificate authority and the "independently" is not a requirement?

[dlongley]

I assume that a DID can be used by a centralized registry, identity provider, or certificate authority and the "independently" is not a requirement?

Correct.

[msporny]

Recommend closing as the charter already does what the issue submitter has requested.

[nadalin]

I disagree with closing as the charter does not do this

[msporny]

I disagree with closing as the charter does not do this

How does the charter not allow for any of the following: "centralized registry, identity provider, or certificate authority"?

[nadalin]

@msporny It's not clear, read my issue and your response, so the centralized registry, identity provider, or certificate authority are optional so suggest

"These new identifiers are designed to enable the controller of a DID to prove control over it may be implemented independently of any centralized registry, identity provider, or certificate authority"

[dlongley]

The statement means: "These new identifiers are designed to be implemented independently of any centralized registry, identity provider, or certificate authority." This is true and a major reasoning behind the design. It does not say that you can't implement them in some other way. We simply talk about what the technology enables here. Note that the other ways may be unbounded -- so we can't and shouldn't enumerate them -- that would lead to even more confusion. So, I think it is clearest as is.

[nadalin]

@dlongley as written, when reading it this sounds like it is only for implementations that do this independently from a central registry, identity provider, or certificate authority. So what is wrong with clarifying it with a MAY, as I'm not asking for any exmaples

[dlongley]

@nadalin, your example text with "may" is not correct, it changes the meaning of what is being said.

The current text says two things (which are the two things we want it to say):

1. These new identifiers are designed to enable the controller of a DID to prove control over it.
2. These new identifiers are designed to be implemented independently of any centralized registry, identity provider, or certificate authority.

I'm not sure where to work in a "may" there that wouldn't make it more confusing or detract from what we're trying to enable. It is not a design goal to enable the implementation of DIDs over centralized registries (etc) -- rather, the design goal is to enable their implementation without those. Again, that doesn't mean you can't implement it with them... existing identifiers already work that way. That's why there is innovation here; we're enabling more.

[jandrieu]

+1 @dlongley The point of the charter is to explain what is being done and why. The WHY is because recent developments in technology allow a new structure for collaborative consensus, making decentralized identifiers possible. These statements are statements of fact, describing the motivation behind the work.

There should be no changes here.

[nadalin]

@dlongley I assume it's not the goal of this WG not to allow or to make this specification useless for Centralized registries (etc), this is what I'm looking for clarity on. I don't think that MAY changes anything but adds the clarity that Centralized registries (etc) are in scope.