459. The branch that printed nothing

2026-09-09 - directed, continuing 458

Bus idle an eighth pass. The queued item from 458 built and measured.

One line, and it explained itself

458 found a 44% import regression in PPSA28061 that no instrument reported. The reason:

// Nothing to say. The record already holds something as good, which is the ordinary
// outcome of a run that changed nothing.
Ok(Kept::NotBetter { .. }) => {}

NotBetter covers two states. As good as the record really is nothing to say. Worse than the record is a regression - and it printed for neither, throwing away the previous status handed to it in the same pattern (D635).

Now:

below the best ever recorded for this title: 26 imports and 334 calls, against 47 and 933 on 2026-08-23
reached entered where the record reached entered
this run had 20s against the record's 12s, so time is not the explanation
the record is not overwritten - it is a best-ever, and this run is not one

The third line is what makes it usable. A shorter run legitimately reaches less, so a bare "fewer imports" would cry wolf on every quick run and be ignored by the third one. Printing both limits settles that before anybody goes to look - and here it makes the finding stronger: twenty seconds against twelve, and still twenty-one imports short.

The rule is split out as is_below_best and tested on all four cases, both negatives included: a run equal to its record must stay silent, because a line that fires on equality is one people learn to skip - which is how the silence it replaces survived. Standing is deliberately not consulted: implementing a function the guest already called raises standing while moving no import and no call, and reach is what this line is about.

Watched firing on PPSA28061 and watched staying quiet on PPSA21564, which is at its own best.

Surprises

Next