Re: WebID default serialization for WebID 2.x

On Tue, Jan 25, 2022 at 6:19 PM Kingsley Idehen <kidehen@openlinksw.com>
wrote:

> On 1/25/22 10:58 AM, Melvin Carvalho wrote:
>
> Someone asked me out of band if this tiny example would work as a WebID:
>
> {
>   "@id": "<webid>",
>    "@type": Agent,
>    "name": "whatever",
>    "public_keys": [a,b,c]
> }
>
> While not perfect, its seems not far off at all.
>
> Imagine if WebID were that simple.  You could use it today, already with
> the public keys it will handle auth, and extensible to many other things.
>
> Hi Melvin,
>
> WebID and WebID-Profile document are indistinguishable i.e., they are
> conflated in your opening sentence.
>
> A WebID is one thing.
>
> A WebID-Profile Document is another.
>
> Speaking about them ambiguously leads to confusion.
>
> If I where writing that message I would have said:
> Someone asked me out of band if this tiny JSON snippet would work as a
> *WebID-Profile*:
>
> {
>   "@id": "<webid>",
>    "@type": Agent,
>    "name": "whatever",
>    "public_keys": [a,b,c]
> }
>
> Answer:
> It would work if it was constructed using valid JSON, rather than a mix of
> JSON and JSON-LD.
>
> Structure Template Revision:
>
> {
>   "id": "<resolvable-identifier-that-denotes-this-agent>",
>    "type":
> "<resolvable-identifier-that-denotes-an-agent-class-in-some-vocab>",
>    "name": "<literal label of this profile doc>",
>    "public_keys": [<list-of-public-keys>]
> }
>
> That will work absolutely fine, without any confusion arising from form
> and function, since that's just an Entity Relationship Graph encoded using
> JSON.
>

Would a fair definition of a valid WebId then be something like: A URI is a
valid WebIdentifier if it dereferences to a valid WebId-Profile describing
the URI with the minimum set of required properties (type, name,
public_keys)?

minor: perhaps type could be inferred if name or public_keys had a range in
the context, or perhaps it would be better to be required for simple
consumption by "dumb" clients.

Nathan