Re: WebID default serialization for WebID 2.x

On Mon, 24 Jan 2022 at 20:13, Jonas Smedegaard <jonas@jones.dk> wrote:

> Quoting Melvin Carvalho (2022-01-24 20:01:21)
> > On Mon, 24 Jan 2022 at 19:03, Jonas Smedegaard <jonas@jones.dk> wrote:
> >
> > > Quoting Kingsley Idehen (2022-01-24 17:32:55)
> > > > A WebID was supposed to be as follows:
> > > >
> > > > A resovable identifier that denotes an Agent, unambiguously.
> > >
> > > [...]
> > >
> > > > A NetID is a resolvable identifier that denotes and entity
> > > > unambiguously.
> > >
> > > Right.
> > >
> > > Subject of this email thread is tied to wether WebID should be...:
> > >
> > >  a) the W3C spec generally for Agent IDs - i.e. exact same as NetID
> > >  b) the W3C spec for Agent IDs usable only where Turtle is spoken
> > >     (others can call their Agent IDs NetID or BingoCards or whatever)
> > >  c) the W3C spec for Agent IDs usable only where JSON(-LD) is spoken
> > >     (others can call their Agent IDs NetID or BingoCards or whatever)
> > >
> > > In other words, do we want WebID simple or easy?
> > >
> > >  * simple enough to be broadly usable
> > >
> > > or
> > >
> > >  * easy usable by constraining scope
> > >
> >
> > There could be a compromise solution, that keeps most people happy.  And
> > that is:
> >
> > Keep JSON(-LD) as a SHOULD, have no serialisation as a MUST, this gives
> the
> > feel of a NetID type spec.  Allow content negotiation to be supported,
> for
> > those that use it
> >
> > Have the examples in the text in JSON(-LD), such that a casual developer
> > can deploy via copy and paste, can deploy to github.io, a static server,
> > jekyll etc.
> >
> > Have a @context to make the examples look easy and attractive, that works
> > with User, Agents, IoT etc.
> >
> > Have a few selected examples that show how to deploy WebIDs for different
> > use cases
>
> How is that a compromise?
>
> My list is
>
>  a) WebID have no serialization as a MUST
>  b) WebID MUST somehow involve Turtle serialization
>  c) WebID MUST somehow involve JSON(-LD) serialization
>
> You describe a).
>
> Remember a few times in this thread I asked specifically if some
> expression had to be read as "MUST" or could be read as "SHOULD"?
>

It's a compromise to me, because I want MUST, but can live with SHOULD,
provided there are nice examples

My sense is that that is something everyone could live with, but I could be
completely wrong :)


>
> I don't mind if WebID SHOULD land on the moon tomorrow and fart darts
> every wednesday - as long as it is not a MUST - because SHOULDs are
> recommendations I can choose to ignore (my applications have no business
> on the moon, and darts are dangerous to kids in my neighbourhood) but
> any MUST ignored means I am violating the spec and can no longer
> reasably describe my application as using WebIDs - they merely use NetID
> or PokerCards or whatever-strings that are not commonly agreed upon in a
> W3C-governed specification.
>
>
>  - Jonas
>
> --
>  * Jonas Smedegaard - idealist & Internet-arkitekt
>  * Tlf.: +45 40843136  Website: http://dr.jones.dk/
>
>  [x] quote me freely  [ ] ask before reusing  [ ] keep private