spine-techbook

Literate programming for technical books treats your prose manuscript as the source of truth, using a tangle pipeline to extract code deterministically and a weave pipeline to generate readable outputs. This ensures code and narrative stay synchronized.

You extract code from prose by running the tangle pipeline to derive source files, then compile and validate them. After validation, the weave pipeline generates reader-friendly multi-format outputs including HTML, PDF, and EPUB.

Managing code snippets across progressive chapters requires a snippet lifecycle that supports introduce, replace, and append operations. This evolves code snippets to accurately reflect changes across the narrative while maintaining dependency ordering.

Checkpoints validate code states in a progressive-build book by freezing behavior at named, validated milestones. They certify that the tangled source code compiles and passes validation at specific stages, guiding progressive development.

Yes, this approach works for maintaining cross-references and concept linking by preserving symbol relationships and dependency ordering within a structured concept graph, ensuring consistency across all generated formats.

Yes, you can use versioned checkpoints to freeze behavior when released. This ensures that validated states remain stable and certified, preventing updates to later chapters from breaking previously established behavior.