Footnotes and *.docx

Hi there, nice to meet you!
I’m trying to compile a document in Scrivener for Windows.
This document has footnotes and I’m trying to compile directly in *.docx.
When I open the *.docx (or *doc, it’s the same) the footnotes are correctly placed and numbered, but if I delete - for instance - footnote number 1 subsequent ones do NOT renumber automatically like I would expect. Footnote number 2 remains number 2 even if footnotes number 1 was deleted. It’s the same for every footnote: if I delete note 30, note 31 does NOT change to 30, 32 does not change to 31 etc.
If I compile in RTF all is perfectly functional, but it would be better if footnotes worked directly in *.docx, obviously. Any suggestion?

This seems to be an issue with MS Word not Scrivener.

Well, one thing to know is that it is better to use RTF with Word anyway, since that is Scrivener’s native format as well as Word’s, so no conversion is necessary. With DOCX, we have to use a third-party conversion utility, so you’re actually putting your content through an unnecessary layer of complexity for no reason. Consider it better for those programs that can’t handle RTF.

That aside, I’m not sure what would make DOCX work but not RTF. I don’t see that on my end, but I don’t have Word and am testing in LibreOffice. The footnotes are dynamic and work fine, there.

Thanks, that’s informative. Bit of a bummer, though, because RTF gives me occasional formatting problems for my intended use. Nice to know, anyway :slight_smile:

RTF hasn’t been Word’s native format for a long, long time – definitely not since Office 2007, which introduced the new XML-based file formats. That said, Microsoft’s RTF to DOCX converter (that is built-in to Word) is definitely a more robust solution.

When I say native, I don’t mean its default file format, but rather that it is capable of loading the format and working with it and saving it “natively”, without internal conversion layers. You seem to be saying otherwise, that it does convert—maybe it does these days, but if it does, it does it so well and completely that we might as well think of it as native support.

RTF wasn’t ever its default format, it was using DOC before DOCX. What RTF was often touted as though is a cross-platform interchange format, for software like Scrivener, to be able to communicate with MS Word without having to reverse engineer DOC.

That need became somewhat less important when DOCX came out and most of the spec was published to the point that anyone could write a DOCX generator or converter (and in fact we do on the Mac side, and should eventually on the Windows side too, to diminish our dependency on the Java converter). But it does remain true to this day: that RTF is meant to be a bridge format between Word and other programs—and to say it is not effectively native to Word kind of defies its purpose.

1 Like

My perspective comes from having worked on some projects for Microsoft (as a vendor) during the time that Office 2007 came out.

The .DOC format was a binary format, like RTF, and used RTF as a starting point. From my understanding, many of the same people who worked on the RTF standard were involved in the design of .DOC – it was intended to be “let’s work around some of deficiencies of RTF without starting over from scratch.” From what I was told, the relationship between DOC and RTF is so tight that a properly written DOC interpreter will do RTF as a side effect.

The .DOCX format was complete re-design from the ground up, based on XML, and as such, the internal workings of modern versions of Word (from 2007 on) no longer speak DOC/RTF internally. The infamous “New Office format” converters Microsoft put out for Office 2003 were a backport of the converters used in Office 2007 to handle the legacy formats. RTF is, as you say, still meant as an interchange format, but it is no longer in any meaningful way a native format for Word and hasn’t been for close to 15 years.

Well I would obviously defer to you on the insider viewpoint, and to the narrower technical definition of a native format. I’m using a broader sense of the term, where one would consider for example PNG to be a native format to Photoshop since it doesn’t require outside or third-party assistance to work with it, despite not being a PSD Photoshop file (its technical native format).

Where I’m coming from is a bit more of a practical, get-it-done, standpoint, where if someone is having difficulties getting .docx output to work correctly for one reason or another and .rtf solves that problem (removing conversion complexity from Scrivener’s side), then it is good to affirm that Word can read this format just fine.

Use what works best, and to that end:

@devinganger: That said, Microsoft’s RTF to DOCX converter (that is built-in to Word) is definitely a more robust solution.

Precisely—I think we’re basically say the same thing here. If Scrivener’s conversion does not work well, then let Word do the heavy lifting.

1 Like