Originally Posted by SBT
uieluck: How much of your workflow is automated with scripts? The more standard the operations, the easier to automate. E.g. stripping of extraneous stuff, do you do that manually?
Personally, I've found it easier to stick the formatting I want into a plain text file, rather than strip out all the cruft word processors stuff in.
- If you've wondered about what a word processor does with your words, think about what a food processor does with your food.
My own workflow is roughly as follows:
- Proofread OCR txt
- Add homebrew markup for headings, footnotes, images etc.
- Add any special-case html code
- run text through the following pipeline of scripts:
- footnote code insertion
- image code insertion
- convert to xhtml
- split xhtml file into chapter files
- generate opf file
- generate ncx file
- repeat relevant steps
- edit opf meta-info
- sometimes edit opf spine
- sometimes edit ncx, renumbering done by script
- edit image size properties in css file
SBT, Yeah I use some scripts and automation tools to help me in the conversion. However, I've never used OCR or Word text. Our books are too stylized and have too many variables.