2026-09-09 - Both magics are accepted, and the cross-census already contained our own file

REQ-20260909T2125Z-3f9d was mostly a retraction of a guess: obSCEne's Makefile defaults EBOOT_GEN to 5 on the Prospero path and passes it to mkself --generation, their eboot leg runs, so it looked as though the measurement already existed. Asked rather than concluded.

It did exist, and twice over:

• build/eboot.bin begins 54 14 f5 ee, built by mkself --generation 5 --privilege root, installed as PPSA99980 and launched - 292 checks ran to completion;
• and the payload leg of the same sweep read it back off console storage: /system_ex/app/PPSA99980/eboot.bin, magic 0xeef51454, ptype 0x1.

The part that is genuinely funny

PPSA99980 is one of the 7 fake/current-magic entries in the 2x2 that settled the magic question two entries ago. The cross-census that proved genuine containers carry the current magic already contained this repository's own output as one of its data points, sitting in the control row, and nobody on either side noticed.

It does no harm - the control row's job is to show the console does not impose the current magic on whatever is installed, and one of the things not being imposed on is ours. But it is a good reminder of how little a container id tells you about what produced a file, which is the same lesson PPSA02664 taught this morning from the other end.

What it changes

Both values are accepted, so the reason the wrap default stayed 4 - that it was the only value shown accepted - is gone.

The default stays 4 on a different reason: it is a route decision. The two proofs are on different delivery routes and neither is on the other's. A package built with the previous magic installs, mounts, loads and executes; a native title directory with the current magic installs and launches. wrap's default serves the package path; the native path passes --generation 5 explicitly and never consults it. Moving the default would put the package path onto a value nothing has accepted there, and D082 already records the two as different delivery routes rather than two spellings of one. (D097)

Three revisions, one surviving sentence

D097 has been rewritten three times today. The population said previous, and that was withdrawn when its evidence turned out to be a fake container. The population then said current, 23 of 23 with a control group - a far better measurement, and also not a reason to move. Now acceptance says both, and the default still does not move, because acceptance turns out to differ by route.

The only sentence that never needed rewriting is the one the first revision added: a default turns on what a loader accepts. Everything that changed was a population, and the population was never the reason.