WebID a system, vs WebID an HTTP URI that ...

[melvincarvalho]

Revisit the terminology section to ensure clear distinction between 'WebID' as an identifier and the broader identity system in: https://github.com/jacoscaz/WebID/blob/main/primary-spec/webid.html

In particular the abstract could be clearer in what "WebID" is and what "a WebID"

The term has multiple uses of course, an ecosystem, a suite of specifications, an HTTP URI etc.

The abstract was quickly put together and we can certainly improve it and clarify terms, as requested in https://github.com/w3c/WebID/issues/37#issuecomment-1879784012

[melvincarvalho]

Current Abstract:

WebID defines a standard means by which user agents and servers interact to establish a user's identity, ensuring a structured, decentralized approach for identity discovery on the web.

How about:

The WebID set of standards, define a means by which clients and servers interact to establish a user's identity, ensuring a structured, loosely-coupled, approach to identity and discovery on the web.

@kidehen is this better? Any suggestions?

[jacoscaz]

/chair hat on

Just to keep everyone aligned, the abstract in my draft now goes as follows:

This specification defines a standard means by which user agents and servers interact to establish a user's identity, ensuring a structured, decentralized approach for identity discovery on the web.

Also, and I'll probably repeat this many times but it bears repeating for clarity, remember that my draft is not meant to replace the 2014 ED, at least for now. It merely aims to provide an integrated view of where the group stands - a representation of consensus, if you will.

/chair hat off

That said, I find that the solution proposed by @melvincarvalho further clarifies things, so +1 from me on it. Quoted below for reference:

The WebID set of standards, define a means by which clients and servers interact to establish a user's identity, ensuring a structured, loosely-coupled, approach to identity and discovery on the web.

[kidehen]

The WebID set of standards, define a means by which clients and servers interact to establish a user's identity, ensuring a structured, loosely-coupled, approach to identity and discovery on the web.

No it doesn't. I proposed leveraging the recursive acronym that's already in the existing spec to reduce friction. Basically, the term "WebID" comprises two things:

1. An identifier
2. An identification system

Remember, a "Name" comprises two things too:

1. Denotatation
2. Connotation

This approach is makes the following both human and machine readable (or computable)

A WebID is an HTTP URI that names an Agent unambiguously. Basically, it is a "Naming System" defined in a coherent spec.

My goal is to use the existing spec to make the above clearer. Once that's achieved, we can then move on to loosening the problematic Turtle dependency.

[kidehen]

This specification defines a standard means by which user agents and servers interact to establish a user's identity, ensuring a structured, decentralized approach for identity discovery on the web.

I can live with that, but I would suggest something along the following lines:

This specification defines a standard means by which an agent is named unambiguously. It addresses user agents and servers interactions for establishing agent identity, leveraging the same architecture that underlies the World Wide Web.

[jacoscaz]

/chair hat off

Ok, let's see:

Option 1., currently in my draft:

This specification defines a standard means by which user agents and servers interact to establish a user's identity, ensuring a structured, decentralized approach for identity discovery on the web.

Option 2., suggested by @melvincarvalho

The WebID set of standards, define a means by which clients and servers interact to establish a user's identity, ensuring a structured, loosely-coupled, approach to identity and discovery on the web.

Option 3., suggested by @kidehen

This specification defines a standard means by which an agent is named unambiguously. It addresses user agents and servers interactions for establishing agent identity, leveraging the same architecture that underlies the World Wide Web.

I do not consider any of these a blocker, though I do have a slight preference towards Option 1. not because it's the one already in my draft but because I subjectively find it clearer and easier to understand. Separately, it might be the one that most of us could live with.

Usual disclaimer: this is note a formal call for voting. I'm just trying to facilitate stating one's preferences.

[kidehen]

For the record, my preferences are as follows:

1. Option 3 -- it takes the a distracting word "decentralized" out of the definition
2. Option 1 -- I can live with that if it emerges as the consensus preference

I cannot support Option 2 because it doesn't address the fundamental issues at hand:

1. Coherent WebID definition
2. Term usage.

Remember, the journey started with "Get yourself a URI" and we want to be able to say "Get yourself a WebID" -- in a meaningful way, to both humans and machines, that's also backed by a simple testsuite.

[jacoscaz]

Closed by #60.