diff options
| author | Ralph Amissah <ralph.amissah@gmail.com> | 2026-05-28 13:16:27 -0400 |
|---|---|---|
| committer | Ralph Amissah <ralph.amissah@gmail.com> | 2026-05-28 13:16:27 -0400 |
| commit | f245b1f010cff053df1192f0788b649158fc6999 (patch) | |
| tree | 8747cc4f0482e9242991742d32efbab6591f016c /org | |
| parent | nix flake update (diff) | |
Diffstat (limited to 'org')
| -rw-r--r-- | org/spine-bespoke-output-homepage-html.org | 23 |
1 files changed, 14 insertions, 9 deletions
diff --git a/org/spine-bespoke-output-homepage-html.org b/org/spine-bespoke-output-homepage-html.org index 7baa474..8e9c5ea 100644 --- a/org/spine-bespoke-output-homepage-html.org +++ b/org/spine-bespoke-output-homepage-html.org @@ -36,9 +36,9 @@ multiple outputs & search</h1> SiSU parses a lightweight-markup source into an abstract document object model. Every substantive element (paragraph, heading, table, verse, image) becomes a typed object carrying its position in the document's sequence -and hierarchy, and a stable citation number. From that single abstraction -it emits multiple output formats - HTML (segmented and scroll), EPUB3, -LaTeX (then PDF via xelatex), ODT, plain text, and an SQLite full-text +and in its heading hierarchy, and a stable citation number. From that single +abstraction it emits multiple output formats - HTML (segmented and scroll), +EPUB3, LaTeX (then PDF via xelatex), ODT, plain text, and an SQLite full-text search database. Each object's number stays stable across every output format and across translations of the same document. </p> @@ -50,12 +50,17 @@ output</b>. <p> <b>Object-Centric Document Abstraction</b>. The abstraction stage builds an -in-memory object model: every paragraph, heading, table, footnote and so on is a -numbered object that carries its own parent / sibling / type metadata, known as -OCN (Object Citation Numbering). Every output format is generated from that -single abstraction, so all formats share the same object identifiers. The -abstraction can also be written out as a human-readable, PEG-parsable text -format (<code>.ssp</code>) that other tools can consume directly. +in-memory object model along two bound dimensions: <b>document structure</b> +(the heading hierarchy - title, section, part, chapter, sub-headings) and +<b>document objects</b> (every paragraph, heading, table, footnote and so on, +as a typed unit). The two are mapped to each other - each object knows its +place in the heading hierarchy, and each structural node knows the objects +that belong to it. Substantive objects also carry a stable citation number +(OCN, Object Citation Numbering). Every output format is generated from this +single abstraction, so all formats share the same object identifiers and the +same structural shape. The abstraction can also be written out as a +human-readable, PEG-parsable text format (<code>.ssp</code>) that other tools +can consume directly. </p> <h2>ℹ - How this differs from a typical "markup → HTML" |