Tinderbox Forum

How to structure a problem

Hi there.

I’m evaluating Tinderbox to solve a problem that I haven’t been enjoying solving in TheBrain. To whit creating a deck of cards based on The 36 Stratagems.

Each of my cards describes a stratagem with an image and different blocks of text. For each card, so far,I have:

  • title(s)
  • summary of action
  • image
  • description
  • list of components
  • list of forces
  • list of questions
  • outcome

The output product should be a CSV file that I will use with InDesign DataMerge to produce the cards for printing.

What I am not clear about is the best way to represent the different chunks of information related to each of the cards.

I had assumed I could attach multiple texts to each note that I could give a name but it looks like string attributes are more intended for “short” text.

Then again, perhaps the right way is to structure this as some kind of “outline of notes” where I have a note for a Card and then connected notes for forces, questions, and so on…

Does anyone have any examples (I’m struggling to find beginner example docs) or guidance about how they’ve structured similar projects?

Thanks in advance.


A few random suggestions:

A sketch is a good starting point for something that needs export like this. Back of a napkin is fine. It helps focus the mind.

Text attributes can contain a lot more than “short text”. Remember, $Text is a text attribute and it can contain thousands of words per note. The limitation is what can be displayed for attributes in Key Attributes listings, not the attribute itself.

There was a recent discussion of CSV export – check this out.

However, CSV will NOT work for images. If you need to export the image consider HTML. (If compatible with InDesign).

Hi Paul.

Thanks for your responses.

My comment about “short text” was more about the editing environment. So far as I can see a string attribute gets only a single line above the notes pane. Or have I missed a way to have a better editing experience for those?

I tried to experiment with a List attribute but that seemed to break the entire attributes interface when I add it to a note. I may well be understanding/doing that wrong.

The export of the image only needs to be an image file name (for InDesign to find the image) not the image data itself. While I’d like to be able to see the images in TinderBox if that was not possible it wouldn’t be a great hardship.

Thanks again.


Not sure what that means. Lists (and sets) are attributes that hold multiple values separated by semicolons. Which is how they are shown in Key Attributes:

I defined a user list attribute type. But as soon as I tried to add it to a note, the attributes part of the UI completely disappears…

But, now you’ve shown me how they’re supposed to work, I can see they’re not what I was looking for.



That might be a bug you’d want to sent to support@eastgate.com offline?

Going back to my original question though:

Are attributes the best way to structure this?

I’ve seen that, from the outline view, I can nest thoughts. It’s slightly unattractive UI-wise at the top-level but I’m wondering if a series of notes each with “Analysis”, “Forces”, “Questions”, and “Outcomes” sub-notes with image perhaps as an attribute mightn’t be a good way of organising this project…?


If you are going to export to CSV then I think attributes are a better choice than child notes.

Each line of the CSV would probably correspond to a card. Each line has multiple “columns” in CSV output, and each column could correspond to the attributes you listed in your original post: title(s), summary of action … and so on.

I’d use child notes for those aspects of each card that require lots of text, you you need more than one. I’d use string attributes for things like the name of the image and the summary of the action, since these will typically be one-liners.

Okay, thanks Mark. It makes sense as a way of arranging the information.

I’m kind of puzzled that all of Paul’s posts seem to have been withdrawn, he seemed to have some useful thoughts & suggestions…

Hrmm… given I will be building 36 notes with the same “internal” structure I wonder if this is what one of prototype, templates, or composites are meant for…

Okay I understand why I seemed to break something. I defined my attributes as:

  • strategic-type
  • strategic-forces
  • chinese-name

and it seems like Tinderbox doesn’t like you putting “-” in your attribute names.

As soon as I changed to:

  • strategicType,
  • strategicForces
  • chineseName

all was well.

If I’ve not also misunderstood something more fundamental about what is going on here it might be an idea to either prevent typing a “-” into the attribute name field or, at least, generate a warning when one does so.


36 notes (or any number of them) with the same structure does indeed lend itself to a prototype, as you intuited.

Since you are exporting, assign the CSV template that you build to the prototype and the notes assigned to that prototype will inherit the template also. Makes it easy to tweak in one location and not in 36.

You are correct, frequently attributes do not like “special characters” especially ambiguous characters such as “-”

You might want consider adding columns to an outline view – one for each of your attributes – so you can see all values for all your special attributes for all notes in one glance. Also, if you copy the notes in column view they are put on the clipboard as a tab-delimited block and can be paste into other programs (Excel, Numbers, Curio, etc.) as a table.

(Sorry for the missing posts – I was cleaning up an old thread and must have been sleepwalking – I cleaned up the wrong thread :frowning_face:)

In addition to this @PaulWalters suggestion (of using columns in outline view, something I do all the time), the OP, @sandbags, might want to check out the long thread from last year about the various possibilities of the Attribute Browser.

The thread is here, How to use the Attribute Browser, and it helps introduce what is to my mind (rather, for my working style) the most powerful single part of Tinderbox.

It seems like Prototypes are about determining the inheritance of attribute values only?

So I looked into setting up a note with the sub-note structure that I needed but when I duplicate the note (e.g. opt+drag) it doesn’t seem to copy any of the internal structure of the copied note.

Am I missing something?

I couldn’t find anything in the guide or help about templates and maybe that’s the answer?



No, they’re about properties of all sorts. (I mean, everything in Tinderbox is fundamentally an attribute, but prototypes apply more broadly than you may think.) They can set: appearance of the note; complexities of its setting (Actions, Rules, Sort, etc); export preferences; just about anything else.

Don’t know if you’ve tried simple Cmd-C/Cmd-V, copy-and-paste. It replicates all the structure etc of the note.

Also don’t know if you’re familiar with the invaluable aTbRef. It’s not a how-to, but it is a look up guide for specifics of all commands etc. If you go to the site below and do a Cmd-F search, you can find almost any command or property you’re looking for.