Just in case anyone’s interested, Eastgate are currently (and “for a very limited time” only!) doing a 51.5%-off discount (for students et al) on their outliner/notetaker/dataholder/braindumper/whatever software Tinderbox. If you’ve been holding back for a price cut from its usual $229 (and assuming you fit the “academic” criteria), now’s yer chance.
I know this has been asked before. I’ve read those posts, downloaded the trial, etc, but I still don’t really get Tinderbox.
I guess my question; Is Tinderbox worth the cost and time to learn how to use it for someone who is already using Scrivener/Devon? I’ve tried to write book-length work before, and this is the first time I haven’t just ground to a halt after about the 2nd or 3rd chapter. I give Scrivener and its remarkable organizational ability credit for this. Since organization and the ability to “see” the whole project and not get lost in it as I work are such big weaknesses for me, I’m always looking for something to help me in this area. Does Tinderbox add real value to this combination for those of you who are more experienced and successful at this than I am?
I qualify for the academic price, so I’m kind of thinking about buying it, but I’m kind of torn. I want anything that will increase my chances of actually writing a full draft of this book, but I hate to waste $100 if it won’t help me.
Rebecca, have you watched the novel-writing video that’s been placed on the Tinderbox site recently? That’s the best very basic introduction to Tinderbox overall (not just for novel outlining) that I’ve seen.
Trouble with Tinderbox – and I’ve been using it since v1 – is that its trial is crippled. The ten or fifteen note limit is daft. An OmniOutliner file, or indeed a Scrivener file, will do just as well with a few notes.
You just can’t see what the app. can do with such a small number of notes. Its strengths, which are remarkable and unique, really don’t begin to show until you’ve got the number of notes up beyond what you can keep in your head at once. Over 100 notes, and you’ll start to feel its advantages.
The Tinderbox file for the book I’m just finishing up contains over 2,500 notes, plus god knows how many links and cross-references. I can pull out all those notes which match any set of criteria I like. All its searches can be “live”, like smart folders. I can find notes with similar content to the current note (bit like DevonThink). I can add as much metadata as I like, on the fly: suppose I want to gather stuff into chapters, all I have to do is add a new “attribute” to the notes, called “Chapter”, and type in the relevant number. Want to find stuff which hasn’t been assigned a chapter? Easy. Set up an “agent” which continually looks for notes where "Chapter’ = “” – i.e., nothing.
Two other great strengths of Tbx: Multiple views of your info. You can have as many notes open as text windows as you like, plus outlines of various levels, map views, you name it. And – the killer for me – you don’t have to commit yourself when you start off. Begin banging stuff down, then let its structure slowly emerge by linking, gathering, drag-and-dropping, colouring, checkboxing and so on.
Tinderbox has weaknesses, too. One of them is that links have no semantic value currently, so although you can have named links, you can’t have, say, a link called “disagrees” then call up a list of everything which “disagrees” with the current note. (This sounds complex but if you use Tinderbox you’ll see what I mean. And I’ve been banging on to Dr Bernstein about it for a long time. I think he’ll implement it one day, but it’s not there yet. If that’s essential to you, for using Tinderbox as a “reasoning engine”, then you’re better off with CMap Tools (free).)
The other weakness is the very limited straightforward export capability. You can do a straight export of the text of selected notes straight out of the box, but anything more sophisticated needs a special “template” which I must say I’ve not looked into very much, but which people up to speed on that aspect can do amazing things with.
I suppose in the end Tinderbox is halfway between a powerful standalone outlining and structuring app., and an incredibly versatile environment with its own programming language – but the latter view of it does require some investment.
I’d describe the learning curve as a long gentle slope with a few quantum jumps in it. I.e., every new thing you learn produces an equivalent reward, with no real end in sight. Some people prefer the steep learning curve – one minute you can do nothing, the next, you can do almost everything – but I have no particular choice.
Personally, I use, and rely on, Scrivener, Devon Think and Tinderbox. I could probably replace Devon Think with, say, Together or Yojimbo or EagleFiler or whatever. I could just about replace Scrivener with Ulysses or TextEdit Projects or something, though I’d really hate to do so. But I can’t think of any replacement for Tinderbox and if I were only allowed one app on my Mac, it would probably have to be Tinderbox.
This was a gripe of mine for ages too, but the latest versions finally have syntax for this! And it is done well, too. Your example would look like:
Set = links.inbound.disagrees.Name
The first word is to indicate a link set generator. Inbound means to look for links pointing to the note (of course, outbound checks for notes which the source note references). The third option specifies the semantic type for the link, in this case “disagrees”. The last keyword selects the attribute you wish to collect from the linked note(s). You can use any attribute. Here “Name” would collect the names all notes linked to this one as disagreeing, and assemble them into a set type attribute. For precise recall, you could have it reference a user attribute instead of Name, which is defined as:
FullName=$Container . $Name
Then other actions could utilise the set to directly reference the linked notes programmatically. But if all you wanted was a human readable list, Name is fine, and the Set attribute could be included as a KeyAttribute. Checking links becomes trivial.
There are other interesting uses too.
In one of my projects I collect a Boolean variable instead of Name. Through a quirk in the set => boolean attribute coercion, I can then test against the “Set” variable for truth-ness and determine whether or not this note is waiting on anything else. In my case, I don’t really care what it is linked too—just whether or not any “waitingFor” links exist. The usage of sets provides a lot of flexibility. Since sets cannot have duplicate entries (part of what makes my Boolean coercion trick work), you can do some interesting things with that too. Or, sets can also be counted and tested numerically, allowing you to run an agent which returns highly linked to/from notes; or orphaned notes without any networking at all.
The demo is about as useful as a 10 document limit DEVONthink Pro demo would be; but then, any other demo model wouldn’t work well either. Thirty days of full usage? Ha. You’d have to live breathe and eat Tinderbox for that to be of any use. But for people wishing to at least see what it can really do, Tinderbox can load large, complicated projects. You just cannot add new notes to it with the demo. You can definitely edit them; play with agents; see how everything fits together; export and work with attributes. It is only the add note feature that is blocked. The exchange website has some good sample documents. Though many are dated and do not showcase the newer features from recent years.
Thanks, Amber. I’ve started using Tinderbox extensively, partly because of endorsements by you and Michael. All this is useful to understand.
Blessings upon your house, Amber. But I don’t know how I’m going to live without my semantics link gripe. (But I’ll find something else to moan about.
I grokked when it first appeared – and very useful too – but I must have missed this syntax. Thanks for pointing it out.
(Now I must go and send an excoriating email to Nathan Matias. I just supervised his dissertation and he didn’t mention this to me, despite knowing of my agonies.)
By the way… for those contemplating trying Tinderbox, you don’t have to get down to this sort of under-the-hood level for it to be really, really useful. This sort of thing is really for malcontents. And, of course, for Amber, who Knows Everything, And Shares It.)