Scrivener to Slideware/Presentations

A large percentage of my job involves using slideware and making presentations. Through the years I have used PowerPoint, Keynote, Beamer and other assorted LaTeX methods, and some HTML based ones such as Deck.js, S5 and Reveal. I have also made PDFs from word processor documents in landscape with large fonts.

I am currently enchanted with Scrivener for the rest of my writing. I realize that I am pushing the boundaries of its intended purpose, but I would like to write my presentations in Scrivener and output them to a slideware in a friction free manner. I would like a minimal amount of interaction after compiling. I know that I can go Scrivener > OmniOutliner > Keynote, but if I decide to make any changes in the Keynote, I would have to go through the whole process. I would like to Compile, and there you are. Any changes, recompile.

Anyone have a workflow like that going? I would like to avoid LaTeX, because life is too short to spend it tracking down cryptic error messages, but If someone has a Scrivener to Beamer workflow that works smoothly, could you give me the details?


Well, you could compile an outline to a text document and import that into Keynote '09 (I’m not sure if it works in the new Keynote 6). But this would only work for text, would still need editing and refinement, and would likely result in a boring set of slides. I think you’d be better off using Scrivener to prepare a handout to accompany your presentation than for creating the actual slide content.

See PigFender’s post here for some excellent tips on what should go onto slides. I could also point you to some academic articles discussing the same thing, but PF’s post is simpler. :slight_smile:

Yes, Scrivener can export to Beamer via MultiMarkdown (very similar to Markdown, but with extended format support like LaTeX and ODT, as well as addditional syntax useful to authors). To do so, you would want to grab the LaTeX support files package from the official MMD page and install that into your texmf folder.

  1. Assuming you have your MMD project ready to compile, you would choose “MultiMarkdown -> PDF” as the compile format (you will need a recognised LaTeX distribution to see that option; if it isn’t there, compile to LaTeX and typeset it normally).
  2. In the LaTeX Options compile pane select “None / Use Meta-Data”.
  3. Now switch to the Meta-Data compile pane and add the following key-value pairs:
    LaTeX Input: mmd-beamer-header

(Here all regular meta-data should reside, such as Title and Author.)

LaTeX Mode: beamer
LaTeX Input: mmd-beamer-begin-doc
LaTeX Footer: mmd-beamer-footer

  1. Compile the .tex file to a working folder and it should be ready to typeset. You can of course examine the boilerplate files in the support package. You probably have stylesheets prepared that you use, and would like to reference from the preamble over the defaults. All of this is flexible, you could supply your own preamble entirely if you wished, but the MMD support package examples are useful for seeing how MMD output and meta-data can be used.

Now as for actually authoring the MMD itself in a suitable way for Beamer output, you’ll need to look that up. I’ve never done it myself. :slight_smile: To my cursory examination however, it appears that setting “Base Header Level” to 2 is best, to bump Scrivener’s outline output to the section level at maximum. This way top level items in the Draft will become presentation sections. Items within them will become [b]\frametitle[/b], and text content within them will be presented on the slide.

Attached you will find my brief exploration of how it works. Included in the Scrivener project is a compiled example along with the outline used to generate it, and the compile settings necessary to do so (if you have everything installed, you should be able to just hit “Compile” and get the same PDF I did).

Note, your browser may download the attachment as a .txt file. Just change it to .zip and extract normally. (129 KB)

Thank you Amber. This should be a big help. It goes way beyond what I was expecting. I will see if it works for me (not technologically, but creatively.)

Nom, thanks for the advice. I have read extensively on presentations and a problem I have with most things I have read is the certainty of the author that there is one true way. Different types of talks, different audiences, and different topics require different approaches. When I was in graduate school and did my first PowerPoint - it was considered hi-tech back then - I was told my choice of dark type on a pale yellow background was a poor choice. I should use yellow type on a dark blue background, because studies had shown that to be best. Five hundred talks later I bet I’d be pretty tired of yellow on blue!

But of course, the speaker is what makes or breaks a talk. The slides are secondary.

No worries. I hope AmberV’s feedback on Beamer helps you with what you need (it’s over my head, but then I have avoided learning MMD and LaTeX).

As for presentations, there is a lot(!) of guff on the internet about what makes a good presentation, most of it based on someone’s opinion and much of it terrible. There are, however, some opinion writers who’s views are consistent with actual research evidence (e.g. Garr Reynolds) or are directly based on research (e.g. Les Posen). But there are also academic authors that have explored communication effectiveness, either of different slide design (e.g. Michael Alley), or use of visual information to support text (verbal or written; e.g. Sweller) and there is a growing body of research evidence on how to use slideware more effectively. There is no one-size-fits-all approach (from memory, Alley emphasises that), but even with the needs of different audiences and different content, there are still more and less effective techniques for using slides to support a speaker.

In the end though, I agree with you 100%: it is the speaker makes or breaks a presentation.

Somewhat offtopic, but has anyone explored alternatives to the standard Powerpoint/Keynote/etc. slide deck? Some mindmapping software, for instance, has a presentation mode that allows easy navigation between branches, rich content at the leaf nodes, and so forth. Standard practice these days is for the presenter to connect their own laptop to the projector, so in theory at least it would be possible to use pretty much any software tool.

On the other hand, it’s also pretty common to make the presentation slides available after the talk, so whatever tool was used would need to be able to output something coherent to a format like PDF or HTML. I can think of a few programs (cough Scrivener cough) that are quite good at multiformat output, though.


Just a guess - but I suspect Curio could be useful.

Or OmniGraffle Pro if you have it. Can read OmniOutliner files, has presentation mode. I’ve never used it for presentations, mind you, Keynote fulfilling all my needs.

I use slides a lot for presentations and am a solid Mindjet/MindManager user. In fact, most of my presentations begin as a mindmap. And I find making slides unecessarily finnicky. BUT I think I would never use Mindjet’s presentation mode to make a presentation. Just not the right thing.


I’ve played with Prezi, and a few years ago toyed with using NovaMind’s presentation tools, but I keep coming back to Keynote (and, to a lesser extent, Powerpoint). With good transitions, you can achieve much the same visual effect, but with a lot more flexibility. As a regular presenter and lecturer, I appreciate the value of having my speaker’s notes on the screen in front of me while my audience sees my slides (Powerpoint can do this, but it amazes me how few people use this feature). There are numerous other useful presentation features in Keynote and Powerpoint that other software lacks (e.g. pressing “b” for a black screen; navigating between non-contiguous slides without showing the audience what you are doing; incorporating images, video and audio; a visual timer; etc).

Wherever possible, I prepare separate handouts rather than a print out of my slides. My slides support my talk, but they aren’t my talk, so without me they can be… difficult to interpret. My handouts usually contain a large chunk of my speaker’s notes (without all the witty spur-of-the-moment jokes and examples that take so long to prepare and practice). Also, I sometimes use images in my presentations that I have permission to use on overheads, but don’t have permission to distribute, so it’s easier to track usage permission if I separate my presentations from accompanying notes.

I have used Scrivener to prepare presentation handouts, although I haven’t yet established a consistent workflow.

A technique I use to generate presentations from Scrivener directly is to write my paper and the key points, etc in markdown and then use deck.js for presentations.

I agree that the speaker makes or breaks a presentation.
My theory is that there is an order of primacy in how we receive and react to information:
1 - body language
2 - voice
3 - content
…by which I mean that what is said (content) is over-ridden by how you say it (voice), and how you say something (voice) can be over-ridden by how you look (body language) when you say it.

Example 1: A message of “that suit looks great on you” will be given a different meaning entirely to the bare content if said with a sarcastic tone.

Example 2: Tell someone “Thank you for your feedback, I’ll definitely take that on board,” and do so in a completely genuine tone. When you start the second phrase (“I’ll definitely…”), violently flip the listener the middle finger.

Wolfie mentions the certainty some commentators have that there is “one true way”. He is right (of course) that the best approach varies by circumstance (as a rule of thumb, you should ignore any advice on presentations that doesn’t require you put the audience’s needs above your own). As Nom points out, though, that does nothing to change the basic fact that some things have been proven both theoretically and empirically to work. One of those is that if you put words on a slide, people will read them instead of listen to you. People may think they can do two things at once but they can’t, not effectively. Audience attention is a difficult thing to maintain at the best of times. If you want them to appreciate the content of your presentation, you need to get your tone and body language in agreement. But before you even get there, you have to make sure they are watching… hence the advice I gave in the post Nom linked to about text content on slides.

I use Scrivener itself for presentations.

The trick is to put everything in the Research folder.

See my 2011 note on the topic, with an attached screen shot.