Import workable footnotes into Scrivener

I fear I do know the answer to my question already but there is some tiny bit of hope left that even as a long time Scrivener user I might just have overlooked something:

Is there any way of importing a text with footnotes into Scrivener and keep the footnotes workable?

What I mean is: When you import RTF with footnotes you will have superscripted numbers in the document and at the bottom you will find the same superscripted numbers followed by their footnote contents. But these technically aren’t real Scrivener footnotes, either inline or inspector, they are just normal text in a smaller font size.

I had hoped—there’s that word again!—that MMD import would work this way. But I found that the export to MMD turns inspector and inline footnotes to proper MMD footnotes but it does not work the other way round.

That’s not how RTF footnotes should import into Scrivener at all - they should come in as either inspector footnotes or inline footnotes, depending on your preferences, so long as they are “true” footnotes in the RTF file. If they are coming in with superscripted numbers, that suggests that the text was only formatted as footnotes in the RTF document and were not true footnotes in the first place, but we’d be happy to take a look at the RTF document causing you problems if you send it to us.

All the best,

Yes, the MMD importer only parses for the meta-data block at the top of the MMD file, and heading levels as indicated by hashmarks (reproducing the document’s structure as a Scrivener outline).

Thanks for your replies.

The RTF footnote problem I asked about actually is just a little aside problem I stumbled over when I tried importing different kinds of files into Scrivener.

I did not know that there could be true and faux footnotes in RTF files. Or to be more precise: That there could be programs with RTF export functions that give you faux footnotes!

My question was not about a single RTF file with faux footnotes but about the RTF export of a number of Markdown programs.

I name names: Byword, Marked, and MultiMarkdown Composer all export RTF with faux footnotes. The only markdown related program I know that does export RTF with proper footnotes is Ulysses. (And for the record: iA Writer as a strict Markdown editor without any “Multi” or “XL” support does not parse footnotes at all.)

And you cannot even really blame them for not having a proper RTF output as they aim to use Markdown as a replacement of the terrible RTF format.

Which directly leads to my real point: I was looking for a way of getting texts into and out of Scrivener with

  • intact structure (not just bigger fonts for a heading but a heading recognized as a heading of a certain level)
  • non-flattened formattings like italics and bold
  • intact content beyond the main text like footnotes and images (at best even with description text)

In short: I was looking for the solution of the BIG PROBLEM not of Scrivener alone but of OS X in general. (Multi)Markdown looked promising, but it is not yet there.

I guess I will have to start a new topic about this in the MMD forum.

Well, faux footnotes are very simple, I can do them right here[1]. Some programs (including Scrivener, in the RTF Compatibility compile option pane) do include options for “flattening” footnotes so that they can be displayed in the many simpler RTF editors that do not support full page layout views.

As far as I’m aware, they are not actually using a Markdown derivative. They have adapted a syntax that has become familiar, to their existing output engine, an engine that predates Markdown—and thus used a non-standard text syntax. All they’ve done is brought their front-end default syntax in line with what a large segment of the community has become accustomed to. The back-end is still using their own technology.

As does Scrivener for that matter, as the MultiMarkdown->RTF choice just uses Fletcher’s utility script, which in turn just uses the UNIX textutil executable that does a very rudimentary job of copying HTML formatting to an RTF file. For a while Fletcher experimented with creating a true RTF converter that worked straight out of the MMD format, but this project was abandoned (and rightly so I feel) to spend energy on the ODT format which is more modern, straightforward to code for and just as capable (the major drawback being that only a few word processors can work with it, but one is the free LibreOffice, which makes that point rather moot since it can turn that into whatever format you need).

Well it depends on what sort of writing experience you are going for, but everything you put into your bullet list is achievable with MultiMarkdown and Scrivener, because every single thing that MultiMarkdown is, is a simple code you can type and read. There is no flattened vs. non-flattened footnotes, just codes that look not terribly dissimilar from the faux footnote I embedded into this post. Granted, if you do that you can’t use Scrivener’s footnote feature, but you don’t need that if the format you are using can describe footnotes using a system that survives any import/export procedure.

The same goes for images. If you define your images with the syntax, then you will always have a 100% accurate transfer between software.

So as I say, you can already do everything you want, but to do so you have to accept the premise that Multi/Markdown &c have stipulated: that using a simple easy to read and write form of markup is the best way at this point in time to convey semantic accuracy across any writing platform, from a Palm Pilot to an iPhone to a PC to a Mac. That means sacrificing the dedicated formatting features of word processors (including Scrivener), when they are used, in most cases.

You can wait for a program that has seamless conversion between formatted text and MMD (or whatever), but why when you can get that right now?

I think another fundamental point here is that these formats are meant to be the nexus format, or the format from which all other formats spring. They aren’t so much meant to be components in a circular chain of formats. There are scant tools for getting a Markdown file out of an RTF or LaTeX file, for instance, and few of them are perfect. This is mainly because of the simplicity of these formats. They do not attempt to address the trillions of ways things can be formatted, but rather just the core “this is that”. So much must be discarded, and some things are so complicated to parse for that they just stay ignored or become flattened. These formats are meant to be all you work in, from start until the very end, when you at last need an output file that works with the rest of the world. Once you make that output format, that’s it. There is no coming back from it. If you need to revise, you revise in the plain-text file and re-produce the formatted document, you don’t revise in the formatted document and then try to convert that massive complexity back into a simple format.

P.S. I don’t mean to denigrate Scrivener’s formatting->MMD features. I in fact use those heavily myself. I use Scrivener’s native footnote feature, as well as its ability to turn images into code. This is mainly because I do all of my writing on a Mac, period. I have no need to ever leave Scrivener and I do 100% of my revisions and edits in Scrivener. So for me it is very much a One Way Only street, with stuff coming out of the compiler and never coming back in via import. I can easily envision even mild adjustments to my workflow that would change that, and if that happened I would gladly switch to pure syntax so that I could use Scrivener in a circular chain of programs with zero data loss.

[1] And that is all there is to it.