Skip to content

Restructure and refine documentation for clarity and organization - #31

Merged
reiase merged 14 commits into
mainfrom
feature/refactor_doc
Oct 19, 2026
Merged

reiase merged 14 commits into
mainfrom
feature/refactor_doc

Conversation

@reiase

@reiase reiase commented Oct 18, 2026

Copy link
Copy Markdown
Collaborator

No description provided.

reiase added 14 commits October 18, 2026 17:08
- Collapse start/, concepts/, why/, community/, and benchmarks/ sections
  by merging or deleting thin pages (161→123 pages per locale)
- Rewrite governance.md to describe the actual single-maintainer model
- Fold methodology.md into benchmarks/index.md, preserving anchors
- Move retired Cluster and daemon design detail into surviving pages
- Add provenance notes to performance analysis pages for unpublished
  local `.data/` archives
- Fix `pvisor-replay` managed run to pass `--config` instead of rejected
  `--spec`
- Update check-docs.py render assertions for the merged pages
Move analysis and research pages out of why/ and design/ into
dedicated lab-notes/ and design/research/ areas, trim the replay
guide by relocating adapter details and usage to design/replay.md
and reference docs, and mark overlaynet host/OCI interception and
job-checkpoint CLI sections as proposals rather than current
behavior.
Move repeated prerequisites, provenance pointers and lab-notes
summaries to their owning pages and link from the rest; add lab-notes
index pages and a check-docs gate that fails on verbatim prose
duplicated across pages.
Delete public SVG charts not cited by any published page and replace
the hard-coded article count with the current pinned value. Add
`referenced_svg_paths` and a `docs.py check` rule so unreferenced public
SVGs fail validation, and gate benchmark methodology notes on documented
sample-size limits rather than fixed P99 claims.

Update relocated lab-notes and replay links, drop the stale provenance
cross-references in the startup benchmark pages, and document the
archived cleanup manifest in `benchmark.new/README.md`.
Tighten phrasing across benchmarks, concepts, design, guides,
and why pages, align the English and Chinese measurement-status
notices, and refresh the translation hashes.
Make benchmark, design, guide and comparison page titles answer
questions readers actually ask, instead of naming internal topics.
Drop redundant scenario table and batch-archival pointers, refresh
translations.json hashes, and update nav labels to match.
Clarify benchmark interpretation, daemon pause/resume wording, and
network policy, threat-model, and overview passages; refresh
translations.json hashes for the edited pages.
Replace conclusion tables and bold summaries with concise motivation
sections that state the decision and measured evidence up front, then
fold per-page conclusions into a short paragraph. Update translations.
Replace per-page design sections with a stated measured quantity and its
user-facing meaning, and drop the Exit-time startup table and its
generated data. Document the three-section page structure in
benchmark/README.md.
Add a docs/README.md section listing the benchmark page structure,
experiment-design, analysis and editing rules that have reached the
site, and link it from the benchmark guide. Reword benchmark index and
cluster-scalability pages to stop describing the pages themselves and
to keep cohort exclusions scoped correctly, and clarify that startup
publishes Ready and Exit CSV columns while user pages cite Ready only.
Document user-page benchmark rules in docs README

Remove the retired cluster-scalability page and its registry
entries, and reword the startup and filesystem coverage notes to
match the new publication rules.
@reiase
reiase merged commit 97475ac into main Oct 19, 2026
16 checks passed
@reiase
reiase deleted the feature/refactor_doc branch October 19, 2026 01:39
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.

1 participant