gnomonlearn
Docs/RFC-0004/Design Q&A

RFC-0004 · Design Q&A

Versioned draft snapshot

RFCs, reference catalogs, and evidence reflect Spec 3b2e3e7a114fb826b1445d39a48756b8fff629cd. Counts and pending/status notes belong to that snapshot. Read the dated project update for later progress.

Source repositories are private during staging. Links marked “repository access required” need access to the linked repository.

← Read the authored RFC

Q: Should aspect precedence be globally fixed or module-scoped?

Status: Open Issue: RFC-0004 §5.2 defines a fixed compiler order by selector specificity, aspect name, and module canonical name; it has no module aspects precedence block. The older answer instead gives each module control over precedence and calls the conflict M04810. RFC-0004 assigns M04810 to a missing extension point warning and M04811 to cross-module precedence conflict. Recommendation: Retain §5.2’s deterministic order and M04811 for a cross-module precedence conflict. Remove the unsupported module-block rule from the Q&A, or define its syntax and precedence effect in the authored RFC with compiler fixtures before adopting it.

Where normative: RFC-0004 §§5.2 and 10 — pending owner decision

Q: Which merge modes are mandatory for M0 and M1?

Status: Open Issue: The older answer names merge_reject, merge_last_write_wins, merge_manual, and merge_union with M0/M1/M2 rollout. RFC-0004 §5.1 instead requires an explicit additive, override, or guarded aspect mode and §5.2 resolves compile-time conflicts; §4 says no merge occurs at runtime. None of the older modes or rollout stages is an authored aspect contract. Recommendation: Treat the §5.1 additive/override/guarded set as the first-release contract and retire the old answer. If runtime conflict handling is intended, specify it separately from compile-time aspect composition and assign its own owner and diagnostics.

Where normative: RFC-0004 §§4–5 — pending owner decision

Q: How should facet inheritance be represented, if allowed?

Status: Open Issue: RFC-0006 §6 item 4 prohibits facet-to-facet inclusion, but the older answer cites nonexistent §6.4 and says aspects are the exclusive replacement. The authored rule instead directs authors needing shared fields to define an entity that includes the facets; RFC-0004 gives aspects a separate push-composition role. Recommendation: Keep the facet-to-facet prohibition and the entity-composition guidance from RFC-0006 §6. Remove the unsupported exclusivity claim, or explicitly revise both RFCs if a different composition model is intended.

Where normative: RFC-0006 §6 and RFC-0004 §§4–5 — pending owner decision