Simplify dataset processing description

README.md

@@ -42,11 +42,11 @@ text = materialize(articles[chunk["article_id"]], chunk)
 
 ## Processing
 
-The Rust pipeline parses wikitext directly into mdhtml's document model, applies Wikipedia-specific cleanup structurally, serializes each retained article once, and computes chunks from the same document without reparsing. It keeps namespace-0, non-redirect, non-disambiguation pages with at least 250 source bytes and discards pages without enough prose after cleanup. Navigation, maintenance, citation, sidebar, and similar non-article templates are removed using Wikimedia template/category metadata; common semantic templates, math, tables, and links are converted directly. Unsupported expansion-dependent constructs may remain as explicit raw wikitext islands.
+Redirects, disambiguation pages, and very short pages are excluded. References, footnotes, navigation boxes, sidebars, maintenance notices, citations, and other non-article material are removed. Links, lists, tables, math, and common templates are converted to `md`; other templates are retained in a readable form where possible.
 
-The chunker prefers high-level heading boundaries and only falls back to lower headings and block boundaries as needed. Footnote references and definitions are removed by this Wikipedia cleanup; the generic chunker also never duplicates footnote definitions across chunks.
+Chunks target about 700 words and prefer section boundaries. Each chunk stores byte offsets into its article plus the headings needed to identify its section.
 
-See `manifest.json` for exact input checksums, schema version, processing parameters, output checksums, and tool source identifiers.
+`manifest.json` records the dump date, settings, checksums, and row counts.
 
 ## Uses and limitations