search

w3c / json-ld-syntax

JSON-LD 1.1 Specification
https://w3c.github.io/json-ld-syntax/
Other
112 stars 22 forks source link

Update use of _IRI_ to obviate _absolute IRI_ #289

Closed gkellogg closed 4 years ago

gkellogg commented 4 years ago

... use IRI reference where appropriate, and replace relative IRI with relative IRI reference.

For #288.

Preview | Diff

TallTed commented 4 years ago

s/Update use if _IRI_ to obviate/Update use of _IRI_ to obviate/

BigBlueHat commented 4 years ago

@gkellogg there's a couple more "absolute IRI" mentions in the "base IRI" definition, but otherwise this looks good.

gkellogg commented 4 years ago

I stick to my point that some readers might wrongly interpret <a>IRI</a> as meaning "either absolute or relative IRI"... But I have no really good alternative to propose...

But, that's how the base RFC defines it, so trying to invent our own terminology can be more confusing, which I think was partly @lo48576's point. In context, it should be clear what the meaning is, and the term definitions (both in the spec, and in their original locations) should provide all the clarity that is necessary.

Clearly, RFC3987 was sufficiently confusing, that the 1.0 spec got it wrong, but I don't think it's up to us to change that.

pchampin commented 4 years ago

Clearly, RFC3987 was sufficiently confusing, that the 1.0 spec got it wrong, but I don't think it's up to us to change that.

Actually, the rest of the RDF 1.1 WG also got it wrong: https://www.w3.org/TR/rdf11-concepts/#note-iris .

@iherman should we make an erratum? (this is not too bad actually, there are only 6 occurences of "relative" and 4 occurences of "absolute".

iherman commented 4 years ago

@iherman should we make an erratum? (this is not too bad actually, there are only 6 occurences of "relative" and 4 occurences of "absolute".

Yes...