36 The Publishing Pipeline: From Source to Finished Work
Guiding Question: What happens between writing and publication?
A writer finishes a paragraph.
The document is saved.
Moments later a website is updated.
A PDF appears.
An EPUB is regenerated.
A presentation reflects the latest changes.
To many authors, this process feels almost effortless.
Yet behind that apparent simplicity lies a carefully orchestrated sequence of transformations.
This sequence is known as the publishing pipeline.
It is the invisible machinery that turns structured text into finished publications.
36.1 From Writing to Publishing
Every publishing workflow begins with a source document.
That source may be written in Markdown, AsciiDoc, reStructuredText, Typst, or another structured text format.
The author writes only once.
Everything that follows is performed automatically.
The manuscript enters the publishing pipeline.
36.2 The Journey of a Document
Conceptually, a publishing pipeline resembles this:
text id="2v0gmx" Write │ ▼ Save │ ▼ Process │ ▼ Render │ ▼ Publish
Each stage has a distinct responsibility.
Writing captures ideas.
Processing interprets structure.
Rendering creates the chosen output.
Publishing delivers the finished work to its audience.
36.3 Metadata
Every publication contains information beyond its visible text.
Title.
Author.
Date.
Keywords.
Language.
Licensing.
Publishing systems collect this information as metadata.
Rather than embedding these details manually into every output format, the author records them once.
The publishing pipeline distributes them wherever they are required.
Metadata quietly keeps publications organized.
36.4 Templates
Templates define appearance without changing content.
Typography.
Margins.
Navigation.
Headers.
Footers.
Colour schemes.
Page layout.
All belong to the template.
This separation allows the same manuscript to assume many different visual identities while preserving its underlying ideas.
36.5 Images and Other Assets
Books and websites contain more than text.
Images.
Diagrams.
Tables.
Stylesheets.
Fonts.
Bibliographies.
All become assets managed by the publishing system.
The pipeline gathers these resources, places them appropriately, and ensures they appear consistently across different publication formats.
The author need not assemble every component manually.
36.6 Cross References and Bibliographies
Earlier we explored automatic numbering and citations.
The publishing pipeline performs these tasks during publication.
Figures receive correct numbers.
Tables update automatically.
Chapter references remain accurate.
Bibliographies are generated.
Every revision benefits from the same automation.
Consistency becomes the default rather than the exception.
36.7 Incremental Publishing
Modern publishing systems often regenerate only the portions that have changed.
A revised chapter updates the corresponding pages.
A modified image replaces its earlier version.
The workflow remains responsive even for large projects.
Authors receive immediate feedback while preserving the advantages of automated publishing.
36.8 Continuous Publishing
Increasingly, publication occurs automatically.
A document saved locally may trigger:
- website regeneration
- testing
- PDF creation
- deployment
- online publication
The boundary between writing and publishing becomes remarkably small.
For many authors, pressing Save is almost synonymous with beginning publication.
36.9 Automation Behind the Scenes
The publishing pipeline reflects a familiar lesson from the previous part of this primer.
Automation removes repetitive effort.
Instead of manually rebuilding every publication, the computer performs predictable tasks with consistency and precision.
The author remains focused upon ideas.
The publishing system manages production.
36.10 Lessons for the Textsmith
The publishing pipeline demonstrates that professional publishing is not a mysterious process.
It is a carefully organized workflow.
Each stage performs one responsibility.
Each component cooperates with the others.
This philosophy should now feel familiar.
Small parts.
Clear responsibilities.
Automation.
Plain text.
The publishing pipeline is, in many ways, the Unix philosophy expressed through publishing.
36.11 Key Ideas
- Publishing pipelines automate the transformation from source documents to finished publications.
- Metadata separates descriptive information from content.
- Templates separate presentation from writing.
- Assets such as images and bibliographies are managed automatically.
- Cross references and citations are generated during publication.
- Incremental and continuous publishing shorten the distance between writing and publication.
- Publishing pipelines apply the Unix philosophy of small cooperating components to the craft of publishing.
In the final chapter of this part, we step back from individual tools and workflows to ask one final question.
Why does plain text continue to matter in a rapidly changing publishing landscape?
There we reflect on accessibility, preservation, openness, and the future of digital publishing itself.