Better Tags for Note Taking and Data Management While You Focus on Searching and Reading

When you search the web, read a book you may create several tens of tiddlers. Normally your focus is on searching and reading NOT note-taking. That results in a crowd of notes later you do not know which is useful and should be kept which is not! Which is correct and reliable which is not, …

One way if to use some tags (perhaps temporary) and later come back to your notes to manage and organize and create the final notes.

@sobjornstad has done some nice work with tgas, see Mosaic Muse — Soren Bjornstad's public notes
When a note is created without tag, a yellow message alert you to take a tag from a list (these tags appear in red)
He also uses tags in yellow color line NeedAttention, NeedExcision, Stub

I thought to have similar workflow with more tags like

NotSure, RawNote, Correct, …

What do you think and what do you propose?

3 Likes

With TiddlyWiki, I ended sticking to a personal convention for using tags in my “main” (and a few smaller, niche) knowledge base wikis. Not preaching it to anyone, just works for me. I use tags starting with an uppercase letter (both single word and multiword) for setting technical wise relationships between tiddlers, so I can build hierarchies (like ToCs). As for “personal” tags having knowledge load (like software names, programming concepts etc), I prefix them with +. This had caused a number of filtering troubles (see my “poltergeist” adventures spread over a few threads), but fortunately, mr Eric Shulman always came up with life saving solutions :+1:

The advantage of this approach is that whenever I have a TiddlyWiki dump, where each tiddler goes to its own separate file, and I have to search these files using third party tools, I can focus on searching over +prefixed tags only.

1 Like

My approach is that notes taken as I read are tagged Reading Notes. For longer works, the primary work is tagged Reading Notes and divisions by chapter/section/etc. are tagged with the top level work.

So, for example, if I am reading “Some Article” there will be an entry by the article’s name, tagged Read Notes. If it is something long, like “Introduction to Economics”, then “Introduction to Economics” will be tagged Reading Notes and each chapter will be tagged Introduction to Economics.

As I extract ideas, or concepts more broadly, these will get entries tagged to the idea rather than as Reading Notes. So, extending the example above, I might create an entry Supply and Demand Curve, tagged Economics with links back to a chapter in Introduction to Economics.

To remind myself to clean up or refactor entries later, I have a few entries tagged Edit Hints.

So, to summarize in a more hierarchical format than I use in practice:

  • Reading Notes
    • Some Article
    • Introduction to Economics
      • Chapter 1
      • Chapter 2
  • Economics
    • Supply and Demand Curve
  • Edit Hints
    • Stub
    • Needs Splitting

This is definitely a topic I’m interested in and think we as a community could have some really exciting conversations about, though I’ve been wrapping my head around all the dimensions of the larger conversation.

I use a modified version of @sobjornstad’s system

  • most card types have a double-side (inspired by the Source/Sink distinction, for example, I have Ideas and Praxis
  • different card types have different colors and icons to rapidly distinguish visually
  • each tag has different viewtemplates appropriate for their type, both for the tiddler and for their streams row body
  • each card type is sorted differently in different locations, and their context menus within streams are unique
  • if a card does not have a type, a warning message appears with several options which will fill in the appropriate field values for the selected type and mark the tiddler a Stub



Stub tags, NeedAttension, and NeedExcision tags appear in the upper right part of tiddlers

Clicking these tags will bring up a confirmation message prompting the user whether they want to remove the tag or not. That allows me to rapidly fill out a stub tiddler from the reference library, by dragging those items up into streams, filling out additional information, and then removing the stub tag without opening the editor.




Elsewhere I have documented how I convert my reading notes into individual tiddlers using AI. I import those tiddlers and modify all their field values at once using the Commander Plugin. They are also tagged with “Note” which causes all these fields to be transcluded in a citation viewtemplate at the bottom of each

In addition, all my notes as I’m taking them are also sorted into one of 5 categories and each of these has different colors associated with that viewtemplate

  • characterization (orange)
  • description (blue)
  • reaction (green)
  • element (pink)
  • concept (yellow)

These particular categories are pragmatic more than they are objective, as they reflect more my understanding of the relationship between different parts of a text rather than, strictly, an attempt to identify them (though often I feel there’s a compelling case for that as well).


May go into this more at a different time, or in a different topic (as the system needs some revision and reevaluation), but I also have a unique viewtemplate for books which allows for rapidly viewing notes in different configurations

From this menu, for example, one can rapidly review all the notes from that book by section

4 Likes

Nothing fancy, but I use the tag .unsorted, and have a list-links macro on my front page or a sidebar tab to process them with better tags later. The . is so that tag appears at the beginning of the tags list when editing. At times I have had a new button in pagecontrols so that I don’t even have to tag them myself.

@Michael1, you might be interested in Bimba László’s plugins:

https://bimlas.github.io/tw5-locator/

https://bimlas.github.io/tw5-kin-filter/

Alfonso

1 Like

I keep meaning to take a look at incorporating these into my workflow, but every time I look become daunted :stuck_out_tongue_winking_eye: