Split document into 2 columns?

Hi all,
Just wondering in Scrivener, how I can split the final document into two columns, just like the attached screenshots show?

Any help would be much appreciated!


What is your final document target? See section 24.13.2 of the user manual if you export to Word processor or final PDF formats. You can also compile without columns and then enable them in your target app if you need a bit more WYSIWYG control, or export via MMD into LaTex or HTML for more automated control.

Hi nontroppo,
Thanks a lot for the help.

I had a look at section 24.13.2 of the user manual and found the answer there :slight_smile:

Could you please elaborate on “You can also compile without columns and then enable them in your target app if you need a bit more WYSIWYG control”? What’s my target app? and how can I enable them in my target app? All I want to do is export the draft to a pdf file.

Thanks again!

If all you want is a PDF file straight from Scrivener then ignore the rest of my comments. :laughing:

What I meant is you could export to, for example, a Word document and then configure the columns in word. Scrivener can do many fundamental layout tasks but if you really need to do some fine manipulation of the final design, then it is recommended to use a dedicated layout tool. But you may find Scrivener’s output is already good enough…

I see, thanks.

Just curious…any difference between a pdf file straight from Scrivener and export via MMD into LaTex, then generate a pdf file from there?

Also, any recommendation on “dedicated layout tools”? Seems you’re very knowledgeable in this field :slight_smile:

Much appreciated!!

Well, you have much more control compiling to MMD and conversion onto LaTeX, HTML or other layout systems. This is because you can use LaTeX templates or CSS (via PrinceXML) to control every pixel of the layout. Because MMD is plain text, you can also transform it easily before conversion. For example, I have two broad output formats for my scientific papers: DOCX for collaborators who only know Word, and LaTeX/HTML to PDF for submissions / comments. From the same Scrivener project I simultaneously spit out each document type without any fiddling; easy to do because plain text is simple to transform.

If you want properly structured figures+captions, need maths, want a bibliography, proper outlined levels in the output, flexible callouts, then MMD is the best way to go IMO.

I have a set of MMD “filters” which do a whole bunch of stuff, for example if I collaborate with an Endnote user, I keep his references in the Scrivener text in the format he needs, but auto-transform them to Pandoc citations during compile so I can keep using my preferred reference manager Bookends while I’m working on his manuscript. The possibilities are endless.

Dedicated layout tools? These split into TWO broad groups — GUI based and programmatic — you already know about programmatic ones (LaTeX / PrinceXML etc.). In the GUI space I used to use InDesign which is the current 226.8kg Gorilla in the room, but many moons ago so don’t have much current experience with them. However you can convert MMD to Indesign ICML to get the best of both worlds! The problem with GUI tools is they are not easy to automate. I like to hit compile, and get what I want without any more fiddle-factor.

Whether you stick with Scrivener direct to PDF or via MMD, you should invest some time in learning the compile process as this is useful to any workflow. See if you can get Scrivener to display as you want, and you may be satisfied. But remember that if you need more flexibility, MMD workflows, especially in soon to be released Scrivener next-generation, are incredibly powerful.

Wow…thank you!

PrinceXML costs $495 for a desktop license and $3800 for a server license so I guess I’ll go down the MMD and Latex route…just to clarify: I need to learn both MMD and Latex, is that correct? Any good resources for learning them? I’ve learned the “simple” markdown in the past but not MMD.

Sorry I’ve got 2 more questions if you wouldn’t mind :smiley:

  1. The office computer runs Windows and I’m wondering if MMD and Latex would run as smoothly as on a Mac? I remotely remember there’re different Latex packages for Windows, is that correct?

  2. Could you point me to the right direction if I want to output documents in East-Asian languages, like Japanese and Chinese? Anything special I need to pay attention to? I mean they’re not only completely different from English, but could be tricky as well coz in some cases I need to generate documents in those two languages that read from top to bottom and from right to left…I’m sure Latex can handle all that but just have no idea about the details — maybe special packages would be required? maybe on top of that, different “special packages” would be required for both Windows and Mac?

Your help would be appreciated! :slight_smile:

PrinceXML is free to use for personal use, as is another very similar tool PDFReactor, both of these use CSS paged media and other web technologies to generate very complex layouts (see the PDFReactor samples). You can learn more about layout using this method here:

Print-CSS Information.

So if your requirements are non-commercial, you could still consider them. I’ve started using PrinceXML, as it is a 4MB small tool, easy to use and CSS is very simple to understand (and huge amounts of tutorials on its use online).

To use Markdown syntax in Scrivener is very straight forward. With Scrivener 3’s new style system this is going to be even easier: you can work with styled text (strong, emphasis, blockquote, admonition etc.), then during compile Scrivener will add the correct prefix/suffix markup (strong, emphasis, >blockquote, [admonition]{.admon}) automatically.

There are two major tools to convert markdown compiled from Scrivener, MMD (traditionally built in to Scrivener) and Pandoc (multi-format convertor). MMD is the simpler of the two, and the default format Scrivener supports. Pandoc also reads the MMD format, and so you can use it directly from Scrivener’s output (and Scrivener 3 is going to support Pandoc format directly). Pandoc is much more flexible, offers better templating (ability to modify the skeleton that converts the document) and critically for me supports bibliographies. To see how to use Scrivener with Pandoc, I have a tutorial available here. Pandoc works on Windows.

If you go the LaTeX route, then the recommendation by Pandoc is to use MikTeX for Windows. My working knowledge of LaTeX is pretty poor, and I find it incredibly cryptic to modify, but there are excellent templates for Pandoc that works very well for my needs. The neat thing about Pandoc is you can add variables in your Scrivener metadata that control the LaTeX templates. For example, in some projects I want Scrivener to cross reference figures, and other want LaTeX to. I added a simple variable to control this directly from Scrivener’s metadata. So the point is you don’t need to “learn” LaTeX in depth if you find a template that works for your needs.

A LaTeX document should in theory be “portable”, it shouldn’t matter what OS you make it on. The caveat to this is fonts, where opentype support can differ across OSes…

Regarding East Asian languages, both HTML+CSS and LaTeX offer support though I have absolutely no experience of this. LaTeX has an incredibly long history of use so there must be good support, but I have no idea about the specifics. HTML+CSS drives the web, and it supports vertical writing mode:

w3.org/International/articl … ical-text/

The wonderful thing about using Pandoc markdown in Scrivener is you are not “locked in” to any particular route, you can output both LaTeX and HTML>PDF as needs require. And with Scrivener 3’s styles and new compile system, you can mix direct compile (Scrivener’s own conversion routes) with Pandoc-compile using different compile presets — the Scrivener project text stays the same!!! Amazingly flexible…

My hats off to you and my both thumbs up too!!! :slight_smile: