Tinderbox and Scrivener

As a former and mostly failed Tinderbox user (I have v5.12) I periodically try to use this excellent but ubercomplicated resource to link into Scrivener – working on my notes and associations in Tinderbox, able to send them to Scrivener in some way e.g. via Dropbox.

I’ve never managed to do it quickly and easily and painlessly – perhaps someone else has?

As there is this kind of friendly collaboration between L&L and Eastgate I’d love to know if someone with deadlines is using this as a good and quick workflow.

Do you mean in a strictly technical sense, or more loosely, such as ‘In what ways can the two applications be made to work successfully together’?

In the technical sense, I believe that the latest version of Tinderbox has the functionality to export directly to Scrivener (although I’ve not used it yet). You may find more information about this in the Tinderbox forums; there are also quite a few threads there that deal with the use of both programmes, and several significant contributors clearly use both (as does AmberV of this parish).

The extremely useful aTbRef web resource created by Mark Anderson, who’s a Scrivener user, is also worth combing. Finally, another Scrivener user, Steve Zeoli, has written a whole series of very helpful Tinderbox posts on his blog that demystify large chunks of the application (but as I recall, though I may be mis-remembering, they don’t deal with Scr. and Tb together). They’re here: http://welcometosherwood.wordpress.com/tinderbox/. Well worth your time absorbing if you’re a Tinderbox struggler.

Definitely - works very well as a two-way street: import Scrivs to Tinderbox and use the Scriv template to export from Tinderbox. I do it all the time (in a workflow that actually starts with Scapple and then goes into Tinderbox - but that’s a less formal procedure).

Incidentally, and a bit off-topic, Tinderbox 6 is coming on apace and shaping up very nicely. There’s a lot of functionality of the programme I still don’t use, but for the stage of pre-Scrivener writing I do in it (basically all project notes and pre-structuring) I think the new version is much more writer-friendly/useful.

Glad to hear TB is becoming more writer-friendly.

Off topic, but I’ve been wondering if it makes sense to buy into the beta program for Tinderbox 6. I’m an occasional TB user who keeps wanting to use the program more heavily, but am put off by the learning curve.


I only started using Tinderbox at the end of last summer and it’s become pretty indispensable in my workflow. For a new project I tend to start thinking out loud in Scapple, then move to Tinderbox so that I can add greater content and text connections, and then move to Scrivener to finish the writing.

I think the steepness of the curve is partly an illusion given by the power users who are used to coding and understandably post to their confreres in that language. It’s actually very easy to start using it at first as a superb outliner, then as a sort-of mind-mapper and only get into the more complex aspects when needed. (In the first flush of a new user’s excitement I wrote some enthusiastic posts about this on my blog - I probably need to edit them a bit now). I’m what they call a `trivial’ user as I still only use a few attributes for each note, and a few exceedingly simple rules and agents (which are basically smart searches). Why it works (for me) is the unique combination of the outlining, the spatial maps and the hyper-links, but I know some power-users who never use the outline …

I signed up for the Backstage a couple of months ago. It’s a wonderful learning experience being on the dedicated Glassboard forum : experienced users find issues with scenarios I could never imagine and Mark, the developer, rewrites to eliminate them. There are updates just about every week. The interface of TB6 is quite different from 5 (e.g tabbed screens) - almost lovely (something you’d never say about TB5 or earlier) and it has made my writing experience better. Having said that, there is as yet no manual for 6 and the implementations of the features are quite different (new popovers and menus) so it might be better for a relative novice to stick with 5 and learn to be more comfortable with that first using Mark A’s amazing Tinderbox Reference document. Being a part of the Backstage is definitely worth paying for though (I bought a $50 ticket) if you have my kind of awed curiosity.

That got my interest. Somewhat off-topic BUT this is about linking TB and Scrivener and a better writing experience, especially when the source material is less than straightforward. Thanks for sharing your experience of TB backstage.

Thank you for your observations.

I’ve done very basic Tinderbox stuff with agents and attributes and such, and I’m currently using it as my main task manager. But I’ve never found a comfortable place for it in my writing workflow. I envision Tinderbox notes as working something like smart (paper) index cards, with the ability to move things around in Map mode but also search based on whatever metadata. Sounds simple, and the advice I’ve gotten from the TBX people has been good, but somehow it’s never really clicked.

I think part of the problem may not be TBX, but just a limitation of computers in general. Even a big screen starts to get pretty crowded once you put an article or chapter’s worth of notes on it, especially if they are large enough to be readable.

And part of the problem may be the “I know if when I see it” nature of writing in general. It’s really hard to know, in advance, when I will need a particular note, and therefore really hard to apply metadata that will be useful three or four months down the road.


I’m a latecomer to this discussion, but it definitely intersects w/ my work needs. I use DEVONThink Pro to store and organize my documents (and I also use a scripted Tag-annotation system that DEVONThink users devised -highly recommended), and I’ve just started using Scrivner for writing – and I love it. I’ve had Tinderbox, but haven’t mastered how to use it – but it seems like a perfect tool for me to use for mapping out a lot of complex research (e.g., segregating and organizing names, dates, associations, etc.).

I’m wondering what’s the best (suggested) workflow for segregating and organizing names, dates, associations, etc. vis-a-vis Scrivner and Tinderbox (though you also mentioned that you use Scapple as part of this workflow, and that might be an easy entry point). I’m still learning Scrivner, but feel much more comfortable working in it than Tinderbox (despite all that I’ve read and watched as far as instruction, I still haven’t grasped it). Could you tell me more about the organizational schema you use? Do you first map things out in Scapple, then port it into Scrivner, and then import that into Tinderbox? Is there a particular forum page or set of instruction (or video intro, which would be really ideal!) that shows how best to accomplish this kind of Tinderbox - Scrivner (DEVONThink) workflow?

Thanks so much for your help… Really appreciate it!

jprint, welcome to Scrivener and this forum.

Others with more experience may disagree, but I suggest that the best way to learn about Tinderbox is to use it.

Initially, I should treat it as a super-charged outliner. If like most of us you’ve used an outliner before, that shouldn’t be too challenging (and as I mention above, Steve Z’s posts on http://welcometosherwood.wordpress.com/tinderbox/, especially his latest one, are an excellent introduction to that usage). Only having mastered Tinderbox in that mode - and, incidentally, the use of its exporting functionality to Scrivener - would I in your position start trying to use it for segregating and organising names, dates and associations as you propose in your research.

Tinderbox’s learning curve may or may not be steep, but there’s certainly quite a lot to master. It is a kind of language which takes time and familiarity to absorb.

Thank you very, very much for this… I greatly appreciate it.

Yes, that’s how I was considering using it as well. Right now, I’m working on a qualitative research project (i.e., a investigative reporting project), and was considering using the model for doing this in the web video, "Processing Qualitative Data with Tinderbox."I have a number of text files for which I need to segregate and organize names, dates, associations (e.g., overlapping government agencies), etc.

The web video, “Processing Qualitative Data with Tinderbox” seems to provide a means by which I can achieve this efficiently, while even automating some of the segregation and organization of this data, e.g., through Tinderbox’s use of agents.

I’ve tried to follow the instructions in that video to explode text and then process the data in the outliner form that you suggested. But I haven’t been able to figure out how to set up certain parts of this process, such as the Attributes, per the instructions in the web video (which seems to use an earlier version of TB).

I’ve already asked for help in the Tinderbox forum (which seems a bit slow), and I’m wondering if you had any ideas / suggestions for how I can seek help for setting up this process in TB 6? (At this point, I’d even pay for some help just to get things going.)

I’ve got more questions, but I feel I need to first understand the basic parts of this before I get in too deep. Thanks so much for your help…again!

I’d very much like to help, but I don’t think that I’m the best person to do so - for several reasons, including the fact that I’m still on the Tinderbox learning curve myself, and also that I have a project of my own, delayed by illness, which unless I hit a few deadlines soon will cause my friends and loved ones to set upon me with murderous intent. :slight_smile:

My recommendation is that you persist with the Tinderbox forum: MarkA there, in particular, is especially knowledgeable, having been responsible for aTBRef. The advice from him is reliable and good.

I’ve never used Tinderbox, but I watched the video and it occurs to me that most of what he does there you could probably accomplish right within scriv.

For exploding the text - just export your document and then Import & Split at paragraphs (or you could manually spllt it as you type, but that’s not as fun).

And then his categorizing his raw data into “buckets” - you can do much the same sort of thing in scriv via labels, keywords, and custom meta-data.

And then his final sort (sorry, I’ve already forgotten the tinderbox names for things) - project search on keywords, labels, and/or custom meta-data as appropriate and save the result as a collection.

I’d suggest sending an email directly to the Tinderbox support address, too. They’re very responsive, and there’s a good chance you’ll get an answer direct from The Man Himself. (Mark Bernstein, the creator of Tinderbox.)


Tinderbox is on my list of things to check out eventually, and while I vaguely knew that Tinderbox could open a scriv file, I didn’t realize it went to this level:

New to v5.11.0, a File menu option allows direct import of Scrivener file data to create a new Tinderbox document:

  • Each text scrivening and folder becomes a Tinderbox note.
  • Image, pdf, and Web scrivenings are currently ignored.
  • Scrivener keywords are imported to a set attribute $ScrivenerKeywords, and the corresponding label colours become the note $Color.
  • Scrivener statuses are imported to $ScrivenerStatus.
  • Scrivener freeform corkboard positions are imported to $Xpos and $Ypos. If Scrivener is not using freeform corkboard positions, Tinderbox uses its own layout.
  • Scrivener custom metadata is mapped to Tinderbox attributes. Because Scrivener custom metadata labels may not be valid Tinderbox attribute names, Tinderbox uses ‘legal’ version of Scrivener’s internal metadata names; the attributes are found in the Scrivener system attribute group.
  • The Scrivener synopsis, if present, becomes the note’s $Subtitle.
  • Because Scrivener values result in many items’ $Color value being set to light colours, the new TBX’s Map preference is set for Outlines to have “Darker colors”.
  • Scrivener internal references are loaded as basic links. The default link type “[InternalLink]” is mapped to “*untitled”.
  • If an external reference link is found, its URL is placed in the note’s $URL attribute. (If several external references are found, only one URL will be imported.

Sounds promising. I presume the current version can still do all that.

I don’t know one way or the other. But I’d recommend reading the release notes for any new Tinderbox version before installing. They have been known to remove functionality.


With the complete Tinderbox codebase being re-written for v6, a number of little-used features have not been re-implemented. Not necessarily dropped, but if there’s no demand likely a few will not return. But these are mostly old, hypertextual, arcana. Inter-app communication, especially Scrivener/Tinderbox/DEVONThink, is not in that bracket.

Tinderbox v6 has a pseudo protocol (tinderbox://) for opening Tinderbox docs at a certain location; the feature is still evolving but should make it easier to cross-link between apps.

A current limitation (I’m sure unintentional) going from Tinderbox to Scrivener is the route is via OPML which supports a limited range of data - plain text and few attributes (i.e. tags/fields). Meanwhile Tinderbox v6’s RTF writing tools are now using current Apple frameworks, allowing better RTF export so that may provide. Tinderbox can export multiple notes as discrete RTF files, but v5’s Nakakoji view (a composite view of many notes) hasn’t yet been re-implemented. I think that’s simply because it’s complicated and a straight re-creatation of the v5 method in v6 isn’t necessarily what’s needed.

If people using the Scrivener/Tinderbox combo - in either/both direction(s) - need different output from Tinderbox I’d recommend them emailing the developer, explaining their needs. Tinderbox v6 is still in its early days so I’m sure input is welcomed.

I think the term 'trivial' re use of Tinderbox is being mis-read. In the context of TB use, it gets stated in terms of being the opposite of 'non-trivial' (difficult/complex). Thus, just because I may use a trivial amount of an app's features doesn't mean what I do [i]with[/i] those feature is of no merit. The trivial/non-trivial use distinction is extremely useful when trying to pin down some issues, as they tend to only show under non-trivial use. Perhaps sanction is read where none was ever intended. :smiley:  

I know a fair bit about Tinderbox but as yet my use of Scrivener is … trivial! (I'm held back by being unable to figure out the complexity of the export to Word; I end up just copy/pasting the composite Scrivener text view to Word which. I know that isn't the intended export method but it's the only one I can get to work as I can't find a detailed enough worked-through export demo).

The best tool for getting your work out of Scrivener to Word (or anything else) is the Compile function. A number of relevant tutorials can be found on our video tutorials page:

As much information about the Compile function as you could possibly want can be found in Chapter 24 of the Scrivener manual.


Thanks, I’ll give that a try. I quite realise that anything with depth takes a while to customise to taste.

Yeah, I need to find the time to give the trial a proper test to see if I can get my head around it, and whether I can do enough with it to justify the price.

It looks like it might be awesome, but I suspect I have a lot to learn first.