#1113: Moved SHACL-SPARQL Rules section into its own shacl12-inference-rules document

[HolgerKnublauch]

Closes #1113

• See the new Inference Rules document rendered online here

• See the new remaining SPARQL document rendered online here

I basically moved the Rules section without changing semantics. A few editorial improvements were made on this opportunity.

There is a broken reference to shacl12-inference-rules from the SPARQL document right now, which I hope will disappear once the new document has been published.

@nicholascar @caribouW3 do I need to change anything so that the new document can be published?

@simonstey you were one of the editors of the SHACL-AF document and I believe you deserve either to be included as one of the authors of Inference Rules, or be mentioned as former editor. Please let me know your preference. Happy to add you as author if you can still identify with the content :)

[nicholascar]

I wonder if Sections 11 - SPARQL Rules & 12 - Triple Rules can be collapsed down into subsections within a Section 11 - Rule Types? This will better match with Section 2's examples of Rule Types.

I don't have a comment on the changes needed for publication yet.

[HolgerKnublauch]

I wonder if Sections 11 - SPARQL Rules & 12 - Triple Rules can be collapsed down into subsections within a Section 11 - Rule Types? This will better match with Section 2's examples of Rule Types.

There are arguments in both directions. I see Triple Rules as sufficiently distinct from SPARQL rules to go into their own top-level section. But I don't mind either way. If anyone else supports this change, I'll do it.

[nicholascar]

I wonder if Sections 11 - SPARQL Rules & 12 - Triple Rules can be collapsed down into subsections within a Section 11 - Rule Types? This will better match with Section 2's examples of Rule Types.

There are arguments in both directions. I see Triple Rules as sufficiently distinct from SPARQL rules to go into their own top-level section. But I don't mind either way. If anyone else supports this change, I'll do it.

Oh yes, I recognise they are entirely distinct but by putting the SPARQL & Triple rule types in a Rule Types section, their distinction is preserved but it just emphasises their being part of an (open ended) set of potential rule types.

[HolgerKnublauch]

I wonder if Sections 11 - SPARQL Rules & 12 - Triple Rules can be collapsed down into subsections within a Section 11 - Rule Types? This will better match with Section 2's examples of Rule Types.

There are arguments in both directions. I see Triple Rules as sufficiently distinct from SPARQL rules to go into their own top-level section. But I don't mind either way. If anyone else supports this change, I'll do it.

Oh yes, I recognise they are entirely distinct but by putting the SPARQL & Triple rule types in a Rule Types section, their distinction is preserved but it just emphasises their being part of an (open ended) set of potential rule types.

Ok I have moved them into a shared section "Built-in Rule Types" now. This had the added benefit of forcing me to think about compliance. I have added a sentence that engines must report a failure when they encounter unknown rule types.

I hope this allows us to proceed for this first version.

[afs]

This would be a good time to rename shacl12-sparql as

"SHACL SPARQL-based extensions"

#1116

[afs]

This does not close #1113.

There is no decision on whether there is a separate top rules doc.
This document has no material, as content or placeholders, for that.

[mgberg]

I agree that that it might be a good opportunity to do this

This would be a good time to rename shacl12-sparql as

"SHACL SPARQL-based extensions"

#1116

[HolgerKnublauch]

I agree that that it might be a good opportunity to do this

This would be a good time to rename shacl12-sparql as
"SHACL SPARQL-based extensions"
#1116

We'll do that in the dedicated ticket #1116. I want to avoid sidetracking discussions that may delay publishing this new document ASAP.

[HolgerKnublauch]

This does not close #1113.

There is no decision on whether there is a separate top rules doc. This document has no material, as content or placeholders, for that.

Right, that topic was a spin-off from the #1113 thread so I have moved it to its own ticket #1196. It shouldn't hold off moving SHACL rules into its own document.

[afs]

This PR makes some unrelated changes SHACL SPARQL(-based) Extensions.
Was that intended? Can they go in directly now?

[mgberg]

Looks good to me, although I'm not 100% sure what Andy's last comment is referring to.

[afs]

Looks good to me, although I'm not 100% sure what Andy's last comment is referring to.

e.g. sh:sparqlExpr around L1674 is a name change.
L1369 is a link fix.
Various others.

shacl-node-expr has a spelling correction.

[HolgerKnublauch]

Looks good to me, although I'm not 100% sure what Andy's last comment is referring to.

e.g. sh:sparqlExpr around L1674 is a name change. L1369 is a link fix. Various others.

shacl-node-expr has a spelling correction.

Yes intentional, some of these were detected with Claude and are IMHO bug fixes and non-substantial changes. If you can live with those, it would be great if you could remove the Request Changes tag.

[simonstey]

Closes #1113

* [See the new Inference Rules document rendered online here](https://raw.githack.com/w3c/data-shapes/issue-1113/shacl12-inference-rules/index.html)

* [See the new remaining SPARQL document rendered online here](https://raw.githack.com/w3c/data-shapes/issue-1113/shacl12-sparql/index.html)

I basically moved the Rules section without changing semantics. A few editorial improvements were made on this opportunity.

There is a broken reference to shacl12-inference-rules from the SPARQL document right now, which I hope will disappear once the new document has been published.

@nicholascar @caribouW3 do I need to change anything so that the new document can be published?

@simonstey you were one of the editors of the SHACL-AF document and I believe you deserve either to be included as one of the authors of Inference Rules, or be mentioned as former editor. Please let me know your preference. Happy to add you as author if you can still identify with the content :)

I indeed completely missed that comment. yes please! thanks a lot

[simonstey]

various minor remarks

[simonstey]

we also should add an additional entry to local-biblio.js ->

e.g.

"shacl12-inference-rules": {
  title: "SHACL 1.2 Inference Rules",
  href: "https://www.w3.org/TR/shacl12-inference-rules/",
  status: "ED", publisher: "W3C"
},

[simonstey]

what those syntax limitations are exactly isn't mentioned/explained in this (new) document any longer, right?

and fwiw, shacl12-sparql/index.html:2008 only "binds" SHACL-SPARQL processors while this document defines a new class -> SHACL inference rule processors
=> As such, nothing normatively requires an inference-rule processor to reject a sh:construct containing e.g. MINUS or BIND(… AS ?this) right?

[simonstey]

this is a duplicate of the normative statement in https://github.com/w3c/data-shapes/pull/1186/changes#diff-cefab4ec6768a5dec43f0c6625a251a899bb7aca4c350010cab57295aa4852c4R1164-R1166 but with a MUST.

suggestion: cross reference back ->

Suggested change

	<p>
	Note that not all implementations are required to implement both of these types for conformance.
	However, a <a>rules engine</a> MUST report a <a>failure</a> is it encounters a rule for which it does
	not implement any <a>rule type</a>.
	</p>
	<p>
	Note that not all implementations are required to implement both of these types for conformance.
	A <a>rules engine</a> that encounters a rule for which it does not implement any
	<a>rule type</a> reports a <a>failure</a>, as defined in <a href="#rules-execution"></a>.
	</p>

[simonstey]

sh:Rules is not defined in https://github.com/w3c/data-shapes/blob/gh-pages/shacl12-vocabularies/shacl.ttl#L1923-L1930 -> this only defines sh:Rule