Reference Notation3

[nichtich]

Notation3 (N3) is a superset of Turtle 1.1 and it comes with its own syntax for graph terms:

{ :a :b :c } :d :e .

In contrast Turtle 1.2 introduces two forms

<< :a :b :c >> :d :e .

and

:a :b :c {| :d :e |} .

I'm sure each of the three forms have different semantics and I'm also sure this will confuse not only me, so at least some non-normative reference to Notation3 should be given.

[gkellogg]

These Turtle forms do not relate to graph literals, or any other kinds of graphs, at least in the current interpretation. That aside, I question if it's appropriate for Turtle (a REC) to reference N3 (a Note). It might be appropriate for N3 to reference Turtle and discuss the syntactic similarities of graph literals and annotations, particularly if/when annotations are added to N3.

[nichtich]

Well, hopefully at least this issue will be found when someone will search for RDF Turtle and Notation3 to find out whether and how { :a :b :c } :d :e . and << :a :b :c >> :d :e . relate to each other. Both reifyed triple and graph term look, swim and quack alike to the naked eye.

[pchampin]

@nichtich I sympathize with your question, but I concur with @gkellogg that it may not be appropriate for a REC to compare itself to an unofficial specification -- why this one in particular?

Also, if this was to be done, the Turtle specification would not be the right place, because the issue is not so much about the concrete syntax, it is about the abstract syntax / data model.

Anyway, at some point we will have to provide some guidance about where to use triple terms vs. where to use named graphs (both are part of the RDF abstract syntax, and both also look, swim and quack alike). This will probably happen in a (non normative) Note (maybe RDF-Primer, or a future "Best practice" document). It may be appropriate, in such a section, to also refer to N3 Graph terms.