Field report from applying the QEP-2 type labels to QuantEcon/skills' open issues (2026-08-13).
The gap: the type group has no honest label for plan/tracking issues — long-lived issues that coordinate other issues rather than describe one piece of work. QuantEcon/skills has five (a development-plan issue per plugin, plus a cross-thread workplan whose tasks are checklists/sub-issues). enhancement mislabels a coordination artifact as a work item, discuss implies an undecided question, and leaving them untyped reads as "not yet triaged" — which under QEP-2's one-type-per-issue-at-triage rule is indistinguishable from the triage never having happened.
Suggestion: either add a core type — e.g. tracking ("Long-lived plan or workplan coordinating other issues") — or state in QEP-2 that coordination issues deliberately carry no type label. The first seems better, since it keeps "untyped = untriaged" reliable, and native sub-issue hierarchies give tooling an easy way to verify the label is being used as intended.
Interim disposition in QuantEcon/skills: the four plugin plan issues are labelled enhancement as the nearest fit; the workplan (QuantEcon/skills#25) is left untyped pending this report's outcome.
Field report from applying the QEP-2 type labels to QuantEcon/skills' open issues (2026-08-13).
The gap: the type group has no honest label for plan/tracking issues — long-lived issues that coordinate other issues rather than describe one piece of work. QuantEcon/skills has five (a development-plan issue per plugin, plus a cross-thread workplan whose tasks are checklists/sub-issues).
enhancementmislabels a coordination artifact as a work item,discussimplies an undecided question, and leaving them untyped reads as "not yet triaged" — which under QEP-2's one-type-per-issue-at-triage rule is indistinguishable from the triage never having happened.Suggestion: either add a core type — e.g.
tracking("Long-lived plan or workplan coordinating other issues") — or state in QEP-2 that coordination issues deliberately carry no type label. The first seems better, since it keeps "untyped = untriaged" reliable, and native sub-issue hierarchies give tooling an easy way to verify the label is being used as intended.Interim disposition in QuantEcon/skills: the four plugin plan issues are labelled
enhancementas the nearest fit; the workplan (QuantEcon/skills#25) is left untyped pending this report's outcome.