LOAD RDF document clarification (description section and definition section)

[afs]

This closes #41.

This PR changes the description section for LOAD to cover the case where the RDF Document is a quads format.

There is a note that encountering quads/named graph when expecting triples is an error.

See issue #41 for discussion about that.

This PR is draft at the moment; revisions of the formal LOAD definition section to come.

---

Preview | Diff

[TallTed]

For future similar... It will be helpful for reviewing changes, if the line breaks can be addressed in a distinct commit, if not a distinct PR.

[afs]

@TallTed The "preview" and "diff" links are useful for that. The WG discussed in the WG meeting of 2024-11-21.

Unfortunately for some reason it is not getting generated (because it was a "draft PR?"). This seems to be an external problem "Timed out after waiting 30000ms" which seems to need an external fix. Hopefully, they will become available.

The changes are restricted to replacing the text within two sections, and are structured as 2 commits (maybe be squashed before merge), one for each section, for ease of review.

[TallTed]

Suggested change

	The `LOAD` operation reads an RDF document from a IRI and inserts triples
	into the graphs of the dataset in the <a href="#defn_graphStore">Graph Store</a>.
	The `LOAD` operation reads an RDF document from a IRI and inserts triples therefrom
	into the graphs of the dataset in the <a href="#defn_graphStore">Graph Store</a>.

[afs]

"therefrom" is archaic at least in the OED (public access).

and declining

😄

Do you have another suggestion?

[TallTed]

Suggested change

	The `LOAD` operation reads an RDF document from a IRI and inserts triples
	into the graphs of the dataset in the <a href="#defn_graphStore">Graph Store</a>.
	The `LOAD` operation reads an RDF document from an IRI and inserts triples from the document
	into the graph(s) of a dataset in a <a href="#defn_graphStore">Graph Store</a>.

[afs]

Reworded to be as neutral as possible. (We can't talk about quads.)

The `LOAD` operation reads an RDF document from an IRI and inserts triples
from the document into the <a href="#defn_graphStore">Graph Store</a>.

Section 2 is "The Graph Store" and this runs through the document.

The destination within the graph store is described later.

[TallTed]

Works for me. I don't have a "resolved" button, but if you do, please press it.

[rubensworks]

Could you also update the changelog at the bottom of this document?

[TallTed]

What is a "slot"? These seem to be stand-ins for "(named) graph". If "(named) graph" can't be used yet, then this should be changed to --

Suggested change

	triples (from a remote graph or dataset) are added to the Graph Store, either in the
	default slot or in a named slot, if specified.
	triples (from a remote graph or dataset) are added to the Graph Store.

[TallTed]

If "(named) graph" can be used, then this should be changed to --

Suggested change

	triples (from a remote graph or dataset) are added to the Graph Store, either in the
	default slot or in a named slot, if specified.
	triples (from a remote graph or dataset) are added to the Graph Store, either in the
	default graph or in a named graph, if specified.

[afs]

"slots" are described in section 2 and 4.2 Auxiliary Definitions.

The text "either in the default slot or in a named slot, if specified." in LOAD is unchanged from SPARQL 1.1 Update in the formal definition section.

A Graph Store is a mutable container of slots (one unnamed slot, and zero or more slots with name). Each slot holds an immutable graph. Update is "replace graph in slot by by another graph", preserving the container.

A dataset is strictly immutable, just like graphs aren't mutable.

The spec is a bit weak in places:

transforms a Graph Store GS to another Graph Store GS'
written
Op(GS, Args) = GS'

but it is the same Graph Store with different contents.

See the note in 4.2 Auxiliary Definitions that mentions to punning on "Graph Store" and "RDF Dataset".

This could be cleared up (but not this PR!).

[TallTed]

My bad. I should have looked to the pr-preview where it's much clearer why "slots" are being discussed.

[afs]

See also : w3c/sparql-graph-store-protocol#24.