Research Evidence Status

The paused PP-W-rows study, its unresolved evidence, and the separate historical M0 outcome

Milestone 0.20 reviewed whether Calor had enough evidence and authority for a three-arm adopter study: ordinary C#, strong protected C#, and protected Calor. It was a planning and evidence tranche, not a compiler release.

Current PP-W-rows Status: Paused

The user paused the separate PP-W-rows study on 2026-09-11. The central pause handoff preserves two consumed invalid/censored attempts and 442 untouched scheduled identities out of 444. Neither the pilot nor confirmation has completed; no benefit or null result exists.

The single total experiment ceiling is USD1,000. Two reservations of USD25.52 each remain held, USD51.04 total. Actual provider charges are unknown, not reconciled or measured spend. The conditionally authorized permanent-retention exception has not been applied. The paused draft checkpoint records unfinished implementation and reviews. There is no permission to resume. Product releases proceed independently; they do not complete the study, alter its protocol or accounting, or authorize more collection.

Recorded Outcome

The formal M0 result is UNADJUDICATED. The maintainer recorded an administrative stop because the prerequisites for a scientific decision were absent:

no consenting independent human methods reviewer;
no committed adopter or approved recruitment;
no accepted scientific resource cap or inquiry deadline;
no provenance-bearing source-task inventory or reviewer schedule;
no completed joint sizing for the cost, completion, safety, and adoption gates.

Missing inputs were not converted into NOT FEASIBLE, zero demand, or a successful implementation.

What M0 Did Establish

v0.18.0 and v0.19.0 predecessor outcomes and retained limits were reconciled.
Plausible cost mechanisms for protected Calor versus protected C#, and protected C# versus ordinary C#, were stated in falsifiable terms.
The legacy held-out harness was found to be a useful implementation reference, but not an isolated or independently authorized acceptance service.
Recruitment after the 2026-08-04 Call W decision and eligible historical task supply remain unknown.
At the M0 closeout, the redesigned PP-W-rows proposal was deferred pending its own buildability result, bounded spend, null-result acceptance, and approval.

M0 did not establish an economic advantage, safety advantage, adopter handoff, or whole-compiler soundness.

Local buildability passed on 2026-09-10. The subsequent reviewed gate closure used the frozen v0.18 release criteria. This deterministic engineering check is not an agent observation. Later task preparation, registration, and instrument work do not establish a study result.

Earlier financial approval, 2026-09-10: USD250 total, pilot only (superseded ceiling). The recorded user decision and historical authorization receipt accepted the registered stopping rules and publication of negative or null results. The quote left the stage choice unselected, so the recorded interpretation used the narrower pilot-only scope. Its BUDGET_NOT_RUN assessment describes an earlier pre-execution hold, not the current pause after two consumed attempts. Neither state is a pilot null/negative or the stage-2 UNDERPOWERED-CARRIED outcome.

The old benefit ledger's epochRun: false and UNDERPOWERED describe a separate historical registration, not the current redesigned attempts. The M0 disposition and its separate human/participant approval boundaries are unchanged. See the dated study chronology.

Release and Work Boundaries

Completing milestone 0.20 did not itself change Directory.Build.props, create a tag, publish a NuGet package, deploy the website, or consume a software version. At the M0 closeout, the current compiler remained v0.19.0. A later ordinary software release may include these documentation changes, but that release is a separate decision with normal build, review, packaging, and publication gates.

The following boundaries remain:

#1284-#1309 are closed as not planned. They were not implemented.
#1311 is independent bounded correctness work.
M0 did not fund or activate #1254 or #1259. Later PP-W financial decisions are separate and do not re-arm M0.
The M0 tranche created no sd-001 epoch, participant recruitment, prototype service, or paid collection. This does not erase the later redesigned PP-W attempts.

Repository Record

The detailed records are:

Use the software changelog for shipped compiler versions. Use this page for the status of the research evidence.