log:notIncludes deliberatly not mentions something about scope?

[phochste]

For log:includes we write as argument modes:

Schema

$s+ log:includes $o+

where:

$s: log:Formula
(Can also be left as a variable to use current N3 document as scope.)
$o: log:Formula

For log:notIncludes such language is missing. Meaning: it is not allowed that the subject can be left as variable to use the current N3 document as scope?

[william-vw]

you are right, we should extend the description of the log:notIncludes there. feel free to create a PR.

[phochste]

@william-vw thanks. but now there seems to be another opion from Jos, that these log:includes and log:notIncludes should not have a scope in the first place.

[josd]

Am fine to have it again like

  // log:includes
  // Schema: $s? log:includes $o+
  // Object may be a concrete formula or the literal `true` (empty formula).
  //
  // Priority / closure semantics (subject-driven):
  //   - subject = GraphTerm: explicit scope, run immediately (no closure gating)
  //   - subject = positive integer literal N (>= 1): delay until saturated closure level >= N
  //   - subject = Var: treat as priority 1 (do not bind)
  //   - any other subject: backward-compatible default priority 1

  // log:notIncludes
  // Schema: $s? log:notIncludes $o+
  //
  // Priority / closure semantics (subject-driven): same as log:includes above.

but then @phochste please change https://codeberg.org/phochste/notation3tests/src/branch/main/tests/generated/log/notIncludes/fail-subjectUnbound-1.n3

[phochste]

Thanks, but I think we need some more insights. I'm creating a new issue out of this. Something is ambiguous in the spec: w3c/N3#225