Skip to content

docs(analysis): one predicate family, six defects - #508

Closed
Sunrisepeak wants to merge 1 commit into
mainfrom
docs/two-layer-predicate-family
Closed

docs(analysis): one predicate family, six defects#508
Sunrisepeak wants to merge 1 commit into
mainfrom
docs/two-layer-predicate-family

Conversation

@Sunrisepeak

Copy link
Copy Markdown
Member

Analysis of the defect family behind 2026.8.25.1 / #506 / #507, and what is still open.

The six defects share one shape. The gap they come from is that this layer comes from the graph and the platform's layer no longer links are treated as one fact — true on Linux where musl is self-contained, false on Darwin where libSystem is both the C library and the kernel interface.

Contains: the measurement behind each of the six, the two CI holes that let ⑥ and ⑦ survive unseen, the coverage matrix with its two empty rows, the two release-chain fragilities, and a proposed fix order.

Pure docs.

…sion is 'who is replaced'

Not a defect list. The six share one shape, and that shape comes from one
gap in the model: 'this layer comes from the graph' and 'the platform's
layer no longer links' are treated as one fact. They coincide on Linux
(musl is self-contained) and do not on Darwin, where libSystem is both the
C library and the kernel interface.
@Sunrisepeak

Copy link
Copy Markdown
Member Author

Folded into #506 — the analysis ships with the fixes it describes.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants