Skip to content

Document the conservation principle as an explicit non-goal #3

Description

@mcornelison

Summary

The spec should explicitly name what A2AL does not instrument: distinguishing skip-because-can't-parse from skip-because-don't-care.

Origin

Independently surfaced by two MoltBook agents during field-testing — fuck_around_find_out (named the unfalsifiable branch: parse-fail vs skip-care is indistinguishable without read-receipts) and nexussim (asymmetric information cost). Field-test sweep Gap 3.

Proposal

Add a short Non-goals note to specs/A2A-Core.md: on a consensus-free platform with no read-receipts, certain adoption questions are uncloseable without violating the protocol's primary purpose. (Paired-format would close the question, but would subsidize the encoded body's parse cost and contaminate the variable.)

Why it matters

Without this non-goal, operators overreach — reading "no engagement" as "rejected" when it may be "tool-blind" or "unparsed." Naming it makes future overreach an immediate spec violation rather than slow methodological drift. Both the maintainer and the field operator hit this (the "English tax" conservation thread).

Good first issue

Documentation-only, no code: draft the Non-goals paragraph. A clean entry point for a first contribution.

Metadata

Metadata

Assignees

No one assigned

    Labels

    documentationImprovements or additions to documentationgood first issueGood for newcomers

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions