Six-task post-fieldtest documentary tidy. No production-code behaviour
changes; 559 tests green at every per-task gate.
T1-T3: form_a.md additions
- §Definitions intro "Three kinds" -> "Five kinds".
- New `### Class — (class ...)` subsection: EBNF carries optional
superclass (0 or 1 per ClassDef.superclass schema), method
signatures with optional defaults; anchored to
examples/test_22c_user_class_e2e.ail (ok 24/2).
- New `### Instance — (instance ...)` subsection: EBNF carries the
(method NAME (body LAM-TERM))* shape; canonical-form CLASS-REF
rule explicitly stated (bare same-module / qualified cross-module);
two examples — same-module abbreviated from
mq3_class_eq_vs_fn_eq_classmod.ail and cross-module qualified
from show_user_adt.ail; bare-cross-module-class-ref diagnostic
named inline.
- §Types `(forall ...)` line extended with optional `(constraints
(constraint CLASS-REF TYPE)+)?` clause; explanatory paragraph
added with no-instance diagnostic anchor; fifth example added
to the Examples block anchored to cmp_max_smoke.ail.
T4: docs/specs/2026-05-13-form-a-default-authoring.md "seven
carve-outs" → "eight carve-outs" at 6 sites contradicting the
§C4(b) compile-time-embed amendment (commit 9fcda8b). Sites:
preamble line 11, §C1 line 170, §C2 line 191, §C3 line 218,
§"Data flow" lines 363 + 374. Post-edit grep returns 4 surviving
"seven" lines (233, 238, 463, 469), all correctly §C4(a)-scope
or arithmetic/future-state.
T5: crates/ailang-core/src/hash.rs:50-57 — delete empty
`#[cfg(test)] mod tests {}` placeholder + 6-line relocation
comment. Tests live in crates/ailang-core/tests/hash_pin.rs
since form-a.1 T5; placeholder served no purpose. hash_pin.rs
still 10/10 passing.
T6: crates/ailang-surface/tests/round_trip.rs — module-level
`//!` and inner `///` docstrings rewritten to the post-T10
four-property framing (parse-determinism / idempotency /
CLI-pipeline-idempotency / carve-out-anchor) instead of the
retired Direction-1/Direction-2 language. Sibling-crate
breadcrumbs added pointing at the other two enforcement points
(roundtrip_cli.rs, carve_out_inventory.rs).
Drift item B2 (audit-form-a "plan file two sites") dropped per
recon verification: docs/plans/2026-05-13-iter-form-a.1.md
contains zero defective "seven" sites; all four hits are
internally scoped to §C4(a), arithmetic, or future-state.
Decision recorded in the journal.
Closes 2 of 3 fieldtest-form-a spec_gap findings (#2 form_a.md
typeclass surface + #3 form_a.md class-qualifier rule for
instance) and 3 of 4 audit-form-a drift items.
Boss-dispatched fieldtest at milestone close. Agent authored 4 .ail
fixtures (factorial smoke + Show user-ADT + user-class Describe with
two instances + two-module workspace) from DESIGN.md + form_a.md only
(no compiler-source reads, no spec-of-milestone reads). All four
fixtures `ail check` ok and `ail run` produces expected stdout.
Findings (1 bug, 1 friction, 2 spec_gaps, 3 working):
- [bug] instance-method-body unbound-var bypasses `ail check` —
forma_3 first attempt called `str_concat` inside the instance
method body; `ail check` returned ok, `ail build` died with the
monomorphise_workspace "unknown identifier" diagnostic. Same
shape at fn-body level correctly fires `[unbound-var]` at check
time. Next: debug skill, RED-first against `ail check`.
- [friction] no `str_concat` primitive. Every realistic Show MyType
body wants string concatenation; the absence forced the agent to
shape examples around bare int_to_str / ctor-arity-1 patterns.
Next: small planner tidy iter wiring `str_concat` symmetric to
`str_clone` and `int_to_str`.
- [spec_gap] form_a.md has zero references to class, instance,
constraint, or method productions — pre-22 surface only. The
form-a-default-authoring milestone made .ail the authoring form,
but the form_a spec doc remains pre-typeclass.
- [spec_gap] form_a.md leaves the class-qualifier ambiguity in
`(instance (class X))` unspecified — bare-class-name vs canonical-
form rule from mq.1 is not documented at the surface level. The
agent picked the bare-name reading from the corpus; the canonical-
form reading is equally plausible from the schema.
Boss-side cleanup at commit time:
- Deleted 8 stale .ail.json sidecars in examples/fieldtest/ (4
forma_* derived by the fieldtester per old dual-form workflow + 4
pre-existing floats_* from the prior fieldtest milestone that
iter form-a.1 T8 missed because the bash deletion loop globbed
only examples/*.ail.json direct children).
- Updated skills/fieldtest/agents/ailang-fieldtester.md to remove
the now-obsolete "generate canonical JSON sidecar" step (Phase 3
rewritten + Iron Law + Reading list + Common Rationalisations +
Red Flags all aligned to the post-form-a single-form doctrine).
cargo test --workspace: 557 passed, 0 failed, 3 ignored (unchanged
from audit-form-a — the fieldtest fixtures are standalone, not
referenced by any test).
Findings deferred to follow-up iters; the milestone close itself is
still clean.
Iter form-a.1 recon surfaced a load-bearing constraint the original
spec didn't anticipate: ailang-core embeds prelude.ail.json at
compile time via include_str! + serde_json::from_str (workspace.rs:417,
main.rs:474), and ailang-core cannot import ailang-surface to switch
the embed to .ail (crate-cycle, documented at loader.rs:9-13).
§C4 splits into two carve-out categories:
- (a) seven subject-matter carve-outs (unchanged, the original list)
- (b) one compile-time-embed carve-out: prelude.ail.json
Acceptance criteria #1, #2 and §T4 carve-out inventory test updated
from 7 to 8.
Roadmap: new [milestone] "Prelude embed: Form-A as compile-time source"
queued to retire §C4(b) by either splitting the workspace loader
(prelude-init moves to ailang-surface) or generating prelude.ail.json
via build.rs. examples/prelude.ail (rendered iter form-a.0) is the
dual-form symptom that milestone exists to resolve. Strike the [todo]
"Author examples/prelude.ail" entry — satisfied by iter form-a.0.
Re-grounding-checked PASS over six load-bearing assumptions (embed
sites verified at workspace.rs:417 + main.rs:474; crate-cycle
constraint real; eight carve-outs all on disk; prelude.ail green
under both roundtrip tests).
Form A (.ail) becomes the sole authoring surface for every program
in the working tree. The seven negative-test JSON-AST fixtures are
the only post-milestone .ail.json files; everything else is rendered
to .ail and the JSON counterpart is deleted (single representation
per program).
Spec decides the four open design questions inline:
- A1: in-process parse via ailang_surface::parse, no build-time
target/ artefacts (would reintroduce a second representation).
- A2: prelude.ail ships as iter-0 pilot, validated by the existing
every_ail_fixture_matches_its_json_counterpart gate.
- A3: per-file deletion cadence; iter 0 is the singular dual-form
window because the JSON counterpart is the witness the CI gate
uses.
- A4: CLAUDE.md and DESIGN.md reworded — canonical vs authoring
forms separated, JSON-AST stays canonical/hashable, .ail becomes
the authoring entry point.
Roundtrip invariant restates from "two forms agree byte-for-byte"
to "parse is deterministic + idempotent under print"; carve-out
inventory test pins the seven JSON-only fixtures against silent
list drift.
Grounding-check PASS over 8 load-bearing assumptions.
Iteration scope: iter 0 = prelude pilot only (render
prelude.ail.json -> prelude.ail; do not delete the JSON yet; do
not touch CLAUDE.md/DESIGN.md/tests yet).
Re-derives the deferred iters 24.2 (class Show + 4 prim instances)
and 24.3 (print free fn + E2E) against the post-mq architecture.
Iter 24.1 shipped at f38bad8 and is unchanged. The supersedes-note
in the spec frontmatter explains the relation to the predecessor at
docs/specs/2026-05-12-24-show-print.md.
Substantive deltas vs. predecessor:
- Dispatch routing through the mq.2/.3 5-step rule; constraint-driven
filter (mq.tidy T1) is load-bearing for show calls inside print's
body.
- 24.2 grows to include the 22b-Show -> TShow/tshow migration
(~14 fixtures + 2 Rust test files) to eliminate post-prelude.Show
ambiguity, analogous to the existing TEq/TOrd convention.
Grounding-check PASS (re-dispatched after two LBA path corrections;
12 load-bearing assumptions all ratified).
Retires the workspace-global MethodNameCollision pre-pass
(crates/ailang-core/src/workspace.rs:547) via a phased three-iter
milestone. Iter 1 canonicalises InstanceDef.class, Constraint.class,
and SuperclassRef.class (extends ct.1's validator); Iter 2 installs
type-driven dispatch (method_to_candidate_classes index + multi-
candidate residual + AmbiguousMethodResolution / UnknownClass
diagnostics) while the workaround still gates real workspaces;
Iter 3 deletes the pre-pass and ships multi-class E2E plus DESIGN.md
sync. Unblocks the deferred milestone 24 (Show + print rewire).
Step 7.5 grounding-check PASS: all 18 load-bearing assumptions
ratified by named, currently-green tests.
24.2 plan-recon surfaced a spec-gap: adding `class Show` to the prelude
would collide with the user-class `Show` declared by 14 test fixtures
under examples/test_22b{1,2,3}_*.ail.json (plus hardcoded "Show" /
"show" assertions in crates/ail/tests/typeclass_22b{2,3}.rs) through
the workspace-global method-name-collision pre-pass at
crates/ailang-core/src/workspace.rs:547.
Three resolution paths surfaced: (A) prep-iter rename Show→Render in
fixtures+tests, (B) inline-rename in 24.2 (makes iter bigger),
(C) defer 24.2+24.3 until the P2 milestone "Module-qualified class
names + type-driven method dispatch" retires the MethodNameCollision
workaround. User picked C.
Spec status flips Draft → Partial; the document remains as historical
context for the future re-brainstorm. Iter 24.1 retains as standalone
runtime infrastructure (`bool_to_str` + `str_clone` heap-Str
primitives are user-callable today; they wait for the deferred Show
instances to find their primary caller). Roadmap P1 entry flips
`[ ] → [~]` with `depends on:` line pointing at the P2 milestone.
Milestone 24 closes the Post-22 Prelude roadmap entry's second half.
Architecture: `class Show a where show : (a borrow) -> Str`, four
primitive instances (Int/Bool/Str/Float — Float included because
float_to_str is semantically unproblematic, unlike Eq/Ord's IEEE-754
issues), and `print : forall a. Show a => (a borrow) -> () !IO` as
a polymorphic free fn following 23.5's `ne/lt`/etc. shape. Two new
runtime primitives (`bool_to_str`, `str_clone`) let Show Bool and
Show Str honour the uniform `(a borrow) -> Str Own` signature without
codegen intercept.
Three iters: 24.1 (runtime + codegen for the two new primitives),
24.2 (class + 4 instances in prelude.ail.json + DESIGN.md anchor),
24.3 (print free fn + positive/user-ADT/negative E2E + DESIGN.md
sync).
io/print_int|bool|float STAY this milestone — redundancy with `print`
ratified as transitional per the 23-spec precedent for ==/eq. Corpus
migration (86 fixtures) queued as separate P2 follow-up.
Grounding-check PASS: all 12 enumerated load-bearing assumptions plus
two additional implicit claims ratified by currently-green named tests
(eob.1 RC-discipline tests anchor 24.1's critical group; milestone-23
mono unification anchors 24.2/24.3's group).
hs.4 surfaced a leak the heap-str-abi spec didn't account for:
heap-Str slabs handed back by int_to_str / float_to_str and
immediately printed via io/print_str leak at scope close under
--alloc=rc. RED tests at e2e.rs (commit 592d87b) pin it.
Brainstorm settled on naming the rule rather than working
around it: Term::Do.args[*] are walked in Borrow position by
uniqueness and linearity passes, symmetric to Term::Ctor.args[*]
being walked in Consume position. The kind itself carries the
ownership default — no per-op field on EffectOpSig.
The rule alone doesn't close the leak; two heap-Str-specific
shape fixes follow (int_to_str / float_to_str gain ret_mode: Own;
drop_symbol_for_binder's App arm gets a Str carve-out symmetric
to field_drop_call's existing one). DESIGN anchor + WhatsNew +
audit-close from hs.5 roll into this milestone.
Grounding-check PASS — all nine load-bearing assumptions
ratified against currently-green tests (with the two RED tests
ratifying the leak as the spec's premise).
hs.2's implementer-orchestrator BLOCKED with an empirical finding: the iter 18d.3 move-tracking and iter 18b non-escape lowering together prevent static-Str pointers from ever reaching ailang_rc_inc/_dec along any shipping execution path. The previously-proposed runtime sentinel guard and the i64 -1 sentinel slot in static-Str globals were both dead by construction.
Amendment is subtractive: remove the runtime/rc.c guard from §Architecture/Runtime layer + §Components/runtime/rc.c (none needed), shrink static-Str globals from <{i64, i64, [N+1 x i8]}> to <{i64, [N+1 x i8]}> (saves 8 bytes per literal), update GEP indices (i32 0, i32 1 → i32 0, i32 0), drop the §Error-handling 'Sentinel collision' subsection, drop the proposed str_field_in_adt_drops_static_str_noop safety-gate test from §Testing strategy (vacuous against natural fixtures), reframe §Architecture/Codegen-4 as 'codegen-level elision invariant, no runtime guard'. Goal/consumer-ABI claims tightened to 'unified along consumer paths, producer + RC sides differ'.
Acceptance: re-dispatched grounding-check PASS. The amendment leaves hs.1 (committed at c56498a) needing a forward-fix retrofit — that is the new hs.2's scope, planned next.
Two builtins ship together: int_to_str (new) and float_to_str (today type-installed but codegen-deferred). Both produce malloc-backed, refcounted Str slabs via new ailang_int_to_str / ailang_float_to_str runtime exports.
Slab layout is {rc_header, len, bytes, NUL}. Static-Str globals migrate to the same packed-struct shape with a UINT64_MAX sentinel rc_header; ailang_rc_inc / _dec short-circuit on the sentinel. From every caller's POV static and heap Str values are indistinguishable, so existing code paths (eq__Str, compare__Str, io/print_str, per-type drop walks) keep working with a single +8 GEP at the three C-API callsites and no change to drop dispatch.
Out of scope: ++ concat operator, Show typeclass + print rewire, length-aware comparison for embedded-NUL Strs.
Single foreign subject (Qwen3-Coder-Next via IONOS). Two blind
cohorts: one sees only a JSON mini-spec, the other only an .ailx
mini-spec; same four tasks. Fairness anchored by a master source
plus a renderer (built on the existing ailang-surface roundtrip
machinery) that projects two form versions from one canonical
content base. Harness shells out to ail parse / check / build / run
and feeds back location-stripped errors so the JSON-pointer
asymmetry doesn't tilt the experiment. Out of scope for v1: a
verdict on Decision 6, a free-choice cohort, repetitions, a second
subject.
Grounding-check PASS on cma.1 (master + renderer + tests).
Recon surfaced that several draft 'moved OUT' sections were universal,
not mode-specific (agent role boundaries, authority over skills, design
rationale, feature-acceptance criterion, when-not-to-delegate). These
apply in any session — interactive or /boss-active — because agents can
be dispatched, design choices can be made, and skills can be edited
outside /boss too.
Tighten the move-list to the three genuinely mode-specific subsections:
Direction freedom, Notifications, Done-state notifications: WhatsNew.md.
Everything else stays in CLAUDE.md. /boss becomes leaner, the universal/
mode-specific boundary is drawn at content that only applies inside the
autonomous loop. Re-dispatched grounding-check: PASS.
Brainstorm assumed all four 23.4-prep commits had shipped to main.
Branch verification (post-iter/23.4 deletion) showed only prep1
(linearity + check bare-name) is on main; prep2 + prep3 were
attempted on the stranded iter/23.4 branch and never landed.
Spec corrections:
- Goal section: prep-commit status block reflects actual main state.
- Architecture: drop §2 (prep2/prep3 rollback) — no rollback exists
on main; section count drops from four to three artefacts.
- Components table: prep2/prep3 row now reads 'abandoned 2026-05-11
with iter/23.4 deletion', not 'shipped, rolled back in 23.4'.
- Acceptance criterion 1: drop the 'prep2 + prep3 are rolled back'
clause.
- Load-bearing assumption 5: now states prep2/prep3 never reached
main (fixes the silent-assumption pattern this spec exists to
prevent — same failure mode the original Spec-23 had).
- Load-bearing assumption 6: cite the cherry-picked main SHAs
(0caaced, 923dd8c) alongside the stranded-branch originals
(8d39f13, aef4ab8).
- Known costs: drop the '23.4 rollback' line item.
Plan-recon for iter 23.4 surfaced that poly_box_proj.ail.json does
not exist in examples/. The three load-bearing regression fixtures
are poly_id, poly_apply, and poly_rec_capture (the latter at
examples/poly_rec_capture.ail.json, driven by e2e.rs::poly_rec_capture_demo).
Three references in the spec corrected. Substantive intent (three
existing polymorphic fixtures regress unchanged under unified mono)
unchanged.
Re-brainstorm of milestone 23 after the original spec retired in
iter gc.1. The original silently assumed polymorphic free fns are
mono-specialised parallel to class methods; iter 23.4 BLOCKED three
times on that seam. Revised spec corrects the architecture by
committing to a unified typecheck-time mono pass (B1) — one
specialiser for all Type::Forall Defs, codegen-time
lower_polymorphic_call retired in same iter. First real run of the
grounding-check agent (Step 7.5) shipped in gc.1: PASS, 15
assumptions ratified.
Discussion-draft, not an approved spec. Captures today's findings:
- Repeated BLOCKEDs on iter 23.4 family were a spec-defect signal,
not a plan-defect signal.
- Today's spec-verification (brainstorm Step 7 + 8) is purely
linguistic — no code-grounded check of load-bearing assumptions.
- Spec-23's 'work the same way as 22b.3' half-sentence was the
exact load-bearing assumption that broke. Three preps spent
patching what should have been caught pre-approval.
Proposes a new agent that, with fresh context between brainstorm
Step 6 and Step 7, extracts load-bearing assumptions from the spec
draft and searches for an existing green fixture ratifying each.
Default on miss: spec discarded, blocked idea moved to roadmap with
'depends on: <missing prerequisite>'.
Eight open design questions left for the next session's brainstorm
to answer.
Anthropic now reserves /plan as a UI command, so the Skill tool refuses to
dispatch it. Rename the project's plan skill to planner, update the symlink
under .claude/skills/, and adjust references in CLAUDE.md, DESIGN.md,
skills/README.md, and the cross-references between brainstorm / implement /
audit / fieldtest / fieldtester. Plan files themselves (docs/plans/*.md)
keep their name — only the skill ID changes.
Three coupled changes to /implement:
- Boss-context offload via single subagent dispatch (target ~100x reduction)
- Iter isolation via branches (eliminates parallel-implementer conflicts)
- Per-iter journals under docs/journals/ replacing the JOURNAL.md monolith
JOURNAL.md becomes docs/journal-archive.md at acceptance time. Workers
stay at Opus 4.7; saving comes from token-volume relocation only.
Three classes of fix from the LLVM-IR audit:
1. != on Float: fcmp une, NOT fcmp one. one is 'ordered and not
equal' (false for nan!=nan, breaks IEEE / user-fixed constraint).
une is 'unordered or not equal' (true for nan!=nan, matches IEEE
and Rust f64::ne).
2. crates/ailang-codegen/src/synth.rs touches the Float primitive at
five sites the original spec missed: llvm_type, builtin_ail_type,
builtin_effect_op_ret, type_descriptor (descriptor 'Fl' to avoid
collision with ADT-name F-prefix), builtin_binop (becomes a
transitional artefact, replaced by type-dispatched helper).
3. Float literals lower as LLVM hex-float syntax (double 0x4014...);
nan/inf/neg_inf as direct SSA double constants at use site (NOT
via the __unreachable__ analogue, which is a terminator
instruction not a value); container slot layout precision
(monomorphisation vs bitcast).
Mainstream-aligns the Float surface: + - * / < <= > >= become
forall-quantified with codegen-dispatch on Int|Float (parallel to
today's == path), instead of the original draft's wort-style
fadd/flt/etc. Adds nan/inf/neg_inf as builtin constants and
is_nan : (Float) -> Bool. % stays Int-only (no fmod yet).
Iteration plan now flags iter 3 and iter 4 as widening passes
that must hold the no-Int-regression line.
User feedback after first draft:
- family/iter not standard terms — milestone/iteration adopted
- ailang-docwriter folded into audit/ rather than left orphan
- explicit "Core rules adopted from Superpowers" section so each
inherited Iron Law is named at the spec level, not just implied
Brummel