Is there any place where I can see example documents (scrivener source + the output) that are produced (a) very quickly and directly through the compile route, or (b) produced more slowly through further editing in a layout-ish program or markdown?
I’m sorry, but my response to your question is “Huh?”
It’s not an either/or question. The Compile Draft output serves as input to the layout program. You can either leave it alone, or deploy the full resources of the layout program to manipulate it to your heart’s content.
Your question is sort of like trying to compare an iPhoto file to a Photoshop image. The Photoshop image could look exactly like the iPhoto file, or it could be completely unrecognizable, depending on the user’s decisions.
Ah, ok. So any sample of the “leave it alone” version? I am curious what, if anything, it does where one might expect things like toc, chapter headings, nested sections, figure titles, cross-references, index.
Hi Sophie,
You might like to look at the User guide from the Help menu, which will explain the options available when you Compile your draft.
In general, Scrivener is not a page layout program, and as such many of the things you have mentioned you would be better off doing in a separate Word processor/layout program after exporting from Scrivener.
But briefly:
Table of Contents, Index - Scrivener will not create these automatically. Something like this you would be better off doing in a Page layout program after you export the contents.
Chapter Headings/Nested Sections: export settings are available to configure what Scrivener will do with folder titles, document stack titles, and document titles (if you do not know the difference between these, I suggest you read the tutorial or user manual). Note that these will not be recognised as headings by word processors, as RTF files do not support headings in this way. They can, however, be setup with a different format (size, font, style). Scrivener also has some in-built chapter/section numbering ability, if you require this (again, check the manual).
Figure titles and cross references I cannot remember off the top of my head. You would be able to make these things work in Scrivener, but not necessarily in a way that would be recognised as such by other programs like Microsoft Word. Again, these sorts of things are usually the kind of thing you would do in a word processor after you have finished writing the content in Scrivener.
The easiest way to get a sample might be to use Compile Draft on one of your own documents. There are a lot of options, but it’s pretty straightforward.
Essentially, Compile Draft will output exactly what you see on the Scrivener screen, except for possible font substitutions specified in the options. It will not generate a TOC, cross-references, or an index. It has limited ability to autonumber chapters, sections, or figures. It has no knowledge of “styles” in the Word sense, though it will preserve whatever formatting you’ve applied (unless you tell it not to).
1: Write in Scrivener.
2: File>Compile Draft to text file.
3: Run layout command in Terminal.
4: That’s it.
All the stuff beloved by tech writers, such as page and section numbering, cross-referencing, and so on, is done more or less automatically, as long as you insert the right markup to start with. Unfortunately, there is some serious learning to do if you do decide to use TEX-based layout. I wouldn’t recommend it to anyone who doesn’t relish a serious technical challenge. But it works well once it’s set up.
Thanks. I know Scrivener is not a page layout program. I was hoping it had the hooks to make bridging that gap easier.
My target is non-fiction, technology writing. I tend to get work-in-progress reviewed quite often by peers (not editors), so I was hoping it would let me do things like
automatically get a moderately readable version out (once set up, no hand tweaks needed each time)
input some tags to be included in a structured output e.g. XML …, …, … or …, or …. This would make the import into a word processor and consequent styling very easy once set up.
Any advice on these? btw, I don’t own the program, am considering moving to Mac in part because of (programs like) Scrivener.
If you’re interested in XML you might want to take a peek at MultiMarkDown. That doesn’t use exactly the XML tags you’re talking about but it does export to XHTML and I believe you can configure it to pass LaTeX or XML/HTML tags through without mangling them.
One point of clarification, the Multimarkdown route is hardly “slow” in the same sense that working an exported file with InDesign is. Granted, if none of the “out of the box” exports do what you want, there will be a little time required to tweak the output generator, but once that is done, exporting via MMD is only a few seconds longer than a regular compile.
To see a working example of an MMD document that is intended for basic “vanilla” export, visit the link suggested above to Fletcher’s site. If you wish to see a Scrivener project that has a custom XSLT for export, go to the FAQ on this forum, scroll down to the bottom, and in fine print you’ll find a download link that includes everything you need. This demonstrates a set-up where the Scrivener document can be exported using the compiler and the “LaTeX” MMD export engine to produce a BBCode document seconds later with no post-tweaking required.
If that sounds a little weird, MMD creates LaTeX files by running a strict DTD XHTML file through an XSLT to produce a .tex source file. The BBCode XSLT goes by the same process, but of course generates something suitable for viewing on this forum. So the “LaTeX” exporter is just a way to hook in to custom XSLT jobs out of Scrivener’s compiler in one shot.
Since you’ll want to be converting one XML format (XHTML) to another, the XSLT should be a fair bit simpler than the ones that are intended to produce plain-text source files.