Re: urn:solid: for prototyping predicates

po 26. 5. 2025 v 10:21 odesílatel Christoph Braun <braun3@fzi.de> napsal:

> Dear Melvin,
> acknowledged.
>
> All those points above are not relevant for "quick prototyping", which I
> thought you were suggesting to use URNs for in this mailing list topic?
>
> Besides:
> URNs are not resolvable by design.
>

Correct, and that is *why* they fit the “just a name” phase so well.

Resolution is deferred until we know the term deserves a Cool URI.

Devs can still mint different URNs for the same concept.
>
> So it can be *one* approach, but it shouldn’t be the *only* one.
>
> I never claimed it to be. I said that there is no need to use URNs instead
> of URIs.
>

Absolutely.  I’m only arguing that we **add** `urn:solid:` to the toolbox,
not replace HTTP URIs.

Given the power of the semantic web, for example owl:equivalentPropertyOf
there is a potential upgrade path, or at least a hint

I suggest starting a new thread (or change topic) to perrue different
patterns in dogfooding, and leave this one for urn:solid discussion, though
I greatly appreciate the feedback :)


>
> URIs do not have to be resolvable/dereferencable (though they should be).
>

Agreed, http URIs *should* be dereferencable.


> This is a different discussion re "In Solid apps, let’s stop using RDF
> terms that 404", and URNs cannot solve that issue.
>

Nicely put.  Each approach by nature will have different trade-offs.


>
> Cheers
> Christoph
>