Revise SHACL 1.2 Rules abstract

[afs]

This PR takes text from the revised SHACL 12. Core abstract as related to SHACL Rules.

This is no more than a slightly better abstract. The final abstract can be written when the document is complete.

• See the revised abstract online here

[liviorobaldo]

I think the abstract of SHACL Rules should not mention shapes. Accordingly, I would shorten the abstract as follows (since it is short, I prefer to include it here rather than proposing the changes directly in "Files changed"):

This document defines SHACL Rules.

SHACL, the Shapes Constraint Language, is a language for describing the structure of RDF graphs. SHACL may be used for a variety of purposes such as validating, inferencing, modeling domains, generating ontologies to inform other agents, building user interfaces, generating code, and integrating data.

SHACL Rules provides inferencing with the generation of new RDF data from a combination of a set of rules and a base data graph. Rules can be expressed as RDF or in the SHACL Rules Language (SRL).

[afs]

I prefer to include it here rather than proposing the changes directly in "Files changed"

That makes it harder to see the changes - the text layout does not correspond to the HTML layout. I have made the change but it would be helpful to have the comparison that the GH UI provides - it highlights the changes in a suggestion.

The text still has "shapes" because of writing out "SHACL".

[liviorobaldo]

Ok, thanks for making the changes. I'll make them myself next time. For me the abstract is fine.

[TallTed]

inferencing has long bothered me, and I am hoping it bothers others here, and maybe we can stop verbing the noun (inference) that was already a nouning of a verb (infer). Maybe we can just use inferring.

Suggested change

	SHACL may be used for a variety of purposes such as validating, inferencing,
	SHACL may be used for a variety of purposes such as validating, inferring,

[afs]

"Inferencing" is a technical term that is used in several fields, such as AI, and is used in the semantic web area.

e.g.
https://blog.okfn.org/2010/08/02/about-inferencing/
#215

Note: the wording for the rules abstract is based on:
https://www.w3.org/TR/shacl12-core/#abstract

[simonstey]

do we then actually need to copy SHACL Cores's abstract 1:1?

[afs]

@simonstey -- I'm not sure which way to read your comment:

Are you saying we should take verbatim the SHACL Core?
Or that we should take none of it?

As pointed out the SHACL Core text emphases validation. It also puts that before the general text.

We don't have, yet, WG text that works for core, sparql, profiles, ui, rules, it is better to take what we can and adapt it for now. For example, it would be restructured to go put the general SHACL description in the main paragraph, then have a final short section to cover the specific document.

That will take awhile. This PR for a slightly better abstract, not the final one, given we know people are reading the live drafts.

[simonstey]

generating ontologies to inform other agents,

I know this is coming from SHACL-core, but: what other agents?!

[HolgerKnublauch]

I picked that term for either humans or software agents such as LLMs. But since these software agents are currently changing on a weekly basis, I wanted to remain vague and not talk about AI agents.

[afs]

I think "agents" will do for now.
What other suggestions are there?

It is a WG task to craft some abstract text that can be used across all the documents. The core abstract is a good start. #729 (comment).

I've created #747 for that.