Tinderbox Forum

Why isn't the notes showing in reference

Ok, this is a bit difficult to explain in words, so pls excuse me if I am not able to articulate this precisely.

I created a new note in a long document by using [[ and so a new note is created. But the text where I have inserted this new note [[ doesn’t do cross-referencing or say backlinking and show me the text in the new note)

Maybe this video will explain better what I mean to say Dropbox - Screen Recording 2021-12-07 at 5.49.38 PM.mov - Simplify your life

If I understand your problem correctly, you’re missing “< >” wrapper that creates a backlink in the new note.
From aTbRef:

**Making a backlink from the link target**

The method can create backlinks from the destination note. [[<that note>]] will create a link to "that note", and then will append the name of this note to the text of "that note" and link the name back to this note. The mark-up needed is used as a wrapper around the entire input string. Thus:

`[[<New Idea>]]`

…will create a new note called 'New Idea' and link to it. If this method is mixed with other syntax variations, like so:

`[[<New Idea|great idea::Recent innovation>]]`

…will create a new note called 'New Idea' and link to it via anchor text 'great idea' and add a backlink with the anchor text 'Recent innovation'.

IMPORTANT: note how the closing '>' comes *at the end* , immediately before the closing ']]' and not before the '|' as might be assumed if all the different parts of syntax were just chained.

More about creating text links with ziplinks method:

1 Like

oh wow! This worked. Thanks

One more thing!

I know there is no blocks in Tinderbox notes.

But is there any roundabout way to link a particular paragraph in the huge note?

Sorry! Pushing my luck! :slight_smile:

Quick search and that’s what I’ve found: Are ziplinks "paragraph aware" in terms of their location within the $Text field?
As far as I know, nothing has changed since then.

1 Like

To make a text link to a specific destination passage:

  1. Select the link anchor.
  2. Make a link and drag it to the left-hand parking space.
  3. Navigate to the destination note.
  4. Select the destination passage.
  5. Drag the links from the parking space and click on the selected text.

But, further to the last, be aware that linking to a whole paragraph somewhere in the $Text in a note with much text doesn’t change much. Link destinations are not marked in any way in the destination note. When following the link in Tinderbox, the target note will be scrolled so the target (paragraph position) is visible in the text area.

I think the Roam analogies are confusing as it is a case of comparing apples and pears. Both are nutritious fruit, but an apple != a pear no matter much we may wish they were the same. Building from such false assumptions of equivalency simply leads to confusion and a tendency to zero-sum (mis-comparisons) with winner/loser outcomes. Thus trying to describe Tinderbox in terms of Roam design concept or vice versa is mostly an exercise in futility.

†. FWIW, Tinderbox in its original design concept envisaged people using small notes so there was no real need to point to a paragraph (or ‘block’) as the target note was likely only one or two paragraphs so the linking wan’t ambiguous.

I fully agree, but there is something you can do to make Tinderbox look somewhat similar to Roam. Check this file:
example.tbx (100.3 KB)
So here’s how it works:
Every paragraph is a new note ( when making this kind of note, remember to set $Prototype to “Note”) - these are your “blocks”. $Name of those notes is the first 5 words of $Text. “Block referencing” (if I understand this concept correctly) can be done by making an alias of your paragraph-notes. Finally, the parent of those notes is a name of, let’s say, an article you write. When you want to see the whole article, just go to parent note and switch to “Preview”.

BTW You can see the “backlinks” of every note using “Links” window (CMD+7).

1 Like


This is a brilliant roundabout. Planning to try this!

How do I set $Prototype to “Note” for the same?

You can find it inside TBX I uploaded yesterday. “Note” prototype has Edict set to: if(!$IsPrototype){$Name=$Text.words(5)}; and Display Expression: $Text;. There is also “HTML item” modified by me to show $Text of notes without their $Name in Preview.

^children(/Templates/HTML page/HTML item)

Interesting, if ugly, use of Tinderbox. I’m intrigued, what is lacking in Roam that we need to do this in Tinderbox rather than just using Roam, which is a different tool. What is roam lacking?

†. Only in the sense that it using the app in a way that runs against design intent. From experience, that means that someone, after hours committed to the method finds it doesn’t quite work and now needs niche feature enhancements that don’t benefit users as a whole. The same experience makes me worry when people try and make App A be like App B rather than simply use App B or get the desired improvements made in App B, which would be a more logical place. IOW, this isn’t about judging people’s right to their own workstyle, but simply reflecting on observing results of making shallow changes to try and make one App like another App. Interestingly, there is often no real need but simply a lazy assumption that a user’s UIs must all work the same way. But, we don’t make screwdrivers like chisels … and for a reason. :slight_smile:

Well said! Personally, I don’t use Tinderbox that way - I’ve tried for a moment at the very beginning of my Tinderbox journey (like a year ago), but I gave up quickly. I agree, It doesn’t make sense.

  1. Subscription model & pricetag.
  2. Web-based - what if one day you wake up only to find that some notes are partially or fully gone - maybe not that often like in the early days of Roam, but it still happens. (some alternatives like Logseq and Athenes lack some important Roam functions)
  3. No E2EE
  4. Development team - too much talk, not enough action. Development is in stagnation for almost a year now.
  5. Limited export options

Thanks. There being so many apps to follow/try out there, it’s useful to hear of the less trumpeted aspects. Not so as to speak ill of any given app, but to get a clearer understanding of the blockages/problems less discussed.

1 Like

@ShiJianhui knows day-to-day RoamWorld better than I. But I think there are some important reasons tied to Tinderbox’s depth:

  • Tinderbox documents can work proactively to organize themselves and to keep your metadata clean. This matters a lot in the long run.
  • Tinderbox has far more options for importing notes, which reduces friction
  • Tinderbox has more flexible export, which may be important for building sustainable workflow
  • Tinderbox notes can inherit properties of related notes, again making it easier to keep large documents organized.
  • Tinderbox offers lots of views, helping you visualize your information space. Where Tinderbox and Roam offer similar visualizations, Tinderbox is often more mature and more salient.
  • Tinderbox has better tools for handling text.

Roam, on the other hand, has its own advantages.

  • Available on all your machines through the Web — not without cost, as noted above, but nice!
  • Easy to get started.
  • Innovated [[ziplinks]], an area in which Tinderbox is catching up but perhaps still behind.
  • Better markdown support takes a weakness (Roam’s text formatting) and makes it a strength
  • Tons of venture capital, and a bigger community of users.

I’m sad to hear that all that venture money (and what has to be a nice revenue stream) isn’t translating to development speed. I wonder what the problem is? But that’s probably best asked over there…

1 Like

An(unintended) degradation of hypertextual (i.e. link) usage I’ve noticed with all this wiki-derived [[…]] stuff, and now in Tinderbox’s ziplink’ method is a false assumption, actually two:

  • that ‘ziplinks’ are a special type of link. They aren’t, but rather a wiki-derived syntax adopted by Markdown and used in Markdown-centric systems as the means of marking up, i.e. creating a link. Of course if you start from only knowing of links as [[…]] then admittedly why would one know better.
  • that link anchor text is (or must be!) a note’s title/name. Meh. We may title (name) notes so as to use them as discrete (unique) specifiers for link targets but the likelihood of that name being a sensible link anchor text is moot. I prefer a richer choice than a note name which often cannot be unique.

In the latter case I think Tinderbox’s implementation trumps more simplistic Markdown-based systems. In fairness Wiki mark-up, from which Markdown borrows, does allow this [[target name|free text label]] though Markdown seems not to offer this (or if it does, hides the documentation well—having googled fruitlessly for it). The latter might explain why those who’ve only ever used Markdown linking before assume the link’s anchor text must be a note name, and are thus stuck with the unreadable prose style that results.

Another, to me, false assumption in the context is speed alone. Getting to a wrong or sub-par solution my be speedier, but is it better? In turn this draws on fetishisation of keyboard-only input. Even if all users were forced to be accurate touch-typists and have ten functional fingers and thumbs, I’m still unconvinced that we all need to work that way. I think some, probably most, links want a degree of deliberate consideration; that is discrete from formalised systems likely using fixed vocabularies where automated action-code linking is an intended outcome.

As a rule of thumb venture capital get’s invested not because those who invest like (or even understand) the ideas and philosophy behind a project but because they want to make more money out of the invest - fast and without any long term thoughts.

1 Like

I fully agree. At first, I loved how [[…]] method of linking let me do it fast, but lately I’ve started to appreciate linking notes with a mouse - it forces you to think harder/deeper about what connects to what and why, and this extra effort pays in the long run.


Glad you answered that!

The almost no learning curve of Roam is also a feature and a bug. The steeper learning curve of Tinderbox probably unintentional might be a feature? Though there are people who are selling Roam learning courses. (which is Hilarious #imho)