Hierarchical tag system

Hi, I would like to achieve a hierarchical tag system. For example, I have a page and would like to tag this page as

  1. data augmentation

  2. text data augmentation

But the second tag is a subtag of the first tag.

Ideally I can only tag the page as text data augmentation, and the logseq can infer that this page is also linked to data augmentation (supertag). But currently if I only tag text data augmentation, then it wouldn’t link the page to the first tag data augmentation. Thus I am wondering if there is an easier and better way to achieve something like this. Thanks!

2 Likes

I think what you may need is namespaces… use a forward slash #[[heirarchy/level1/level2]]

5 Likes

Thanks! Now I can see the tag in the page of hierarchy. I am wondering what is the best workflow? Should we later only work on hierarchy/level1 instead of level1 page? Since level1 is, in the eye of logseq, not the same as hierarchy/level1.

1 Like

#[[hierarchy/level1/level2/level3]] is not equal to nested tag. For instance, when I need to find tag[[hierarchy]], logseq only links the tag[[hierarchy]], not include tag[[hierarchy/level1]], tag[[hierarchy/level1/level2]] and tag[[hierarchy/level1/level2/level3]], that is what I want.

1 Like

Where can I find info about this in the logseq docs? I’ve searched high and low and can find no docs on this tags/hierarchy/thingy!

  • This feature is called namespaces.
  • It doesn’t have particular documentation.
    • Can find videos online.
    • It is simple enough: just put a forward slash / between a “surname” (left part) and a “name” (right part)
      • Repeat for as many surnames as needed:
        • almost always none
        • sometimes one
        • rarely more than one
  • It is meant to be used for disambiguation of otherwise identically named pages.
  • Using namespaces for hierarchical systems is a well-known bad idea.
  • For a general hierarchical structure, the outliner is enough.
  • For composite hierarchical systems, the right tool is properties.
1 Like