Tags vs page/link

Such a simple solution, but is there a list of of these tricks? Thank you.

1 Like

Hmm, well that would be quite nice if that existed, I’d say!

1 Like

Ok, code ticks. :+1:

1 Like

I think the proper term is back ticks. (I don’t want to misinform anyone from what I typed above.) Is that right @Cito ?

1 Like

I prepend a backslash \ to a hex color code #379432 and it hides the code from the from the pane as expected.

A back tick as I understand it, is `` as per Empty inline code backticks (``) leak their style to the text below it seems to do something different.

If so, it would seem our list is coming together :wink:

A “back-tick” (aka “back quote” or “left quote”) is simply the single left quotation mark that looks like the french acute accent. In Markdown they are used to surround inline code, that’s why I called them “code ticks”. For hex color codes, you would use them like this: `#abcdef`. The error you mention seems to happen with empty back-ticks only which you would normally not use anyway.


My rule is backlink for anything I want to annotate or write about, tag for everything else (actions to take, classification/filter tool).

I like @dmc’s idea about a single term having the capacity to be a concept and tag. I see using [[Microsoft]] as a page catchall for stuff I need to annotate and write about, as opposed to #microsoft being pegged to the company mentioned in the context of something which is not fundamentally important.


Tags now to me seems to be a second level citizen in Obsidian: for local graph of a note, notes with the same tag will not be shown with whatever depth you have set.


Interesting point! Never thought of that.

Now that you point out that discrepancy between tags and links, I may experiment with using links instead of tags in situations where I want that local graph behavior that tags don’t provide.

What I mean is assuming you are linking bidirectionally or at least to [[NoteActingAsTag]] from relevant notes, you could search for something like:

[[NoteActingAsTag]] AND [[AnotherNoteAsTag]]

and you will get the same tag functionality explained above…

“A tag of ssl will find many notes. A tag of aws will also find many notes. But ssl and aws will only find the notes where I’ve configured SSL on AWS resources.”

However, although I can’t vouch that it is a huge difference, according to the current online help document it will not be as fast of a search:

tag: will search for your specified tag within a file, for example tag:#work. This is faster and more accurate than searching for the tag in plaintext #work, as it uses the cached information and ignores text in code blocks and sections that aren’t markdown text.

My apologies if anything I shared here is incorrect, but I know I used to use a similar method to this before my workflow drastically changed.

Thanks for the heads up.

1 Like

As Obsidian improves, I am altering the methods I use to appropriately access the information and ideas in my documents.

With tags, the recent addition I am appreciative of is the nested tags feature. At first, I didn’t see how I could use them to improve my process. Now, I use them everyday. Here is an updated example, to my last post, of how I use tags, compared to links:

The tag I use is: To/Read/Recommended
The link I use is: [[Recommended Books#To Read]]

Once I read a recommended book, and decide I want to start note-taking, I will then transition that book from a TBR tag to a TBR link. My goal is to eventually have zero tags in my tag pane.

Here it is, in practice, with a recommended author I am currently exploring:

  • I create an inbox note with the filename of 20210126084613 North with the Spring, which is the current timestamp and the name of the book
  • I enter Tags: To/Read/Recommended in the front matter of that inbox note, add in my workflow tags and quick contextual notes, and then archive the file
  • when ready to do a quick read, I click the To/Read/Recommended tag, in the tag pane, retrieve the 20210126084613 North with the Spring file from my inbox archive, and then read the book
  • when finished with the first read, I access the file 20210126084613 North with the Spring, update it with new workflow tags and the new location, erase Tags: To/Read/Recommended, and archive the file again
  • I create a page, for the book, with a filename of North with the Spring (1951), and then place a link to that page under the heading [[Recommended Books#To Read]]

The distinction here is that books that get a TBR link will be reread and further examined. These books will remain in my vault, whereas books that get a TBR tag may not get read, at all, and just deleted as I work through my tag listings.

Okay, so that post turned out to be more involved than I intended. I’ve benefitted from the information other Obsidian users have shared, so hopefully this post will be of benefit to someone else.


@dmc Though I use tags in a completely different way, I am currently also considering the option of using nested tags - still haven’t come to a conclusion yet if they would rather structure or hinder my workflow; thus, I was curious reading your practice example. However, to be honest, I haven’t understood yet how nested tags help you with that. Maybe I didn’t get your point exactly - but how does it make a difference for your described workflow if you use a nested tag like #To/Read/Recommended or just a single tag like #ToRead or #Recommended-book?

This occurred to me today, and it’s suddenly put it all in place:

  • Tags are best thought of as “specially designated search terms”. You can search for “recommended” but then you get every occurrence of that word. If you search for #recommended, then you only get the tags. But it helps me to think of the question “Do I want this to be a tag rather than a bidirectionally linked word/phrase?” as equivalent to the question: “Do I want to be able to quickly see the search results for this?”
1 Like

I use a #howto tag and then add a separate #cook or #paint or #photography tags.


How do you tell literature and evergreen notes apart in the graph?

So your experience is that it’s beneficial not to try do decide on a clearly defined set of topic tags, but rather add whatever tags seems most fitting in the moment. Going for a more “impulsive” decision is what eventually give the Zettelkasten its voice to talk back to you?

1 Like

Basically, this is my experience so far, yes. Because it prevents you to stay trapped in your predefined categories of thinking (that would be somehow the same as using folders, where you have to decide from the beginning what is the one topic this note belongs to; direct links - though of course I use them too - actually have the same effect: you have to decide from the beginning where this story might end up!) and makes it possible to discover relations you wouldn’t have seen otherwise.

Most of my productive writing results from using tags in this way.


Thanks! I’ve lately felt some frustration over how I use tags, feeling that I use synonyms too often as one example. Before I stumbled on this tread, my plan was to try to define a fixed set of tags and just use them.

But the discussion here makes it obvious that a combination of clearly defined relationships (that is links between pages) and more vague (tags) is probably best for serendipity when using the Zettelkasten during output.

The links might then function as bridges between different thought patterns in the form of tag clusters.


This is why I’ve been trying to start a conversation about implementing a toggle feature that would allow nested tags to be viewed in the graph as orthogonal trees - around which notes are clumped. I believe this can help us get a clearer visualisation of note structuring.
From what I’ve understood, nested tags are uni-directional, while note linkages are bi-directional. I think, having the above-mentioned scheme could enable to spot hierarchies that we personally default to.


I noticed there are two main practices on using the tags: for holding meta-data, or for list topics/keywords.

The way tags are displayed in graph view make them not well-suited for holding meta-data of notes – the point of graph view is to show the connection of concepts/notes, yet their location in the YAML header makes them meta-data de facto. Add to that, a word prefixed with # makes it automatically a tag. I felt it would great if there is another category, say meta to hold the meta-data, while leaving tags and # for topics.

That is where dataview plugin comes in because dataview is able to search through the meta, but no graph views. I use type for my meta, such as todo, place, event, term, etc. tags are for topics for me.

As to tags vs links, tags for things that I don’t intend to create a page for and links for things that I intend to create a page for.