Page vs Tag Use Case

This is not really a bug but i’m thinking how to use it more effectively.

Tag has a similar function just that it cannot put a space in between. I aware people use the page as connection inline in the sentence.
I think the page can replace the tag for the use case purpose.

For example,
I think everyone should [[learn to code]].
The coding teaches us how to think #[[learn to code]]

Anyone mind to share your opinion?

3 Likes

Yeah this is an interesting question, I think a lot of it comes down to personal preferences and concistancy of one’s own system.

I like what you did here:

That could be a good way to use it, if it flows with the text the using [[this format]] works great. And if it’s related but doesn’t form part of the sentence then #[[this format]]` works great.

Additionally I use #this when it’s just one word, I think it looks really clean and simple.

personally I don’t use #[[this format]] but using it in the way described above seems to be a good workflow

5 Likes

This is a somewhat late reply, but that is because I only joined Logseq in May ’23 and am now starting to get a bit of a feel for the app, thoughI am still very much a newbie.

I am posting the answer below because it may be of use to other newbies, and perhaps some veterans (says he with tongue in cheek :rofl:) .

Regarding this topic, I have seen the question about the difference between hashtags and double square brackets answered with 2 observations:

  • use the square brackets in the middle of a sentence for a smooth reading flow, and use a hashtag at the end of a sentence;
  • “it is pretty much a question of personal preference”

Both make sense, but there is another reason why in some circumstances one may prefer a tag: with the Tags plug-in one can look at a list of tags in use, and where each one is used. IMO it is important not to let the creation of new tags get out of hand, so that list may be of great help. Before creating a new tag, I consider if one of the existing tags cannot be considered as a synonym for the one I am thinking of creating. If it is a synonym, then I will use the existing tag. BTW, in multiword tags I always use hyphens to connect the words into 1, as it were.

Regarding the pages referenced by double square brackets, one can of course get a list of those too but it is a longer, less smooth process than just calling up the tag list.

I don’t know if anyone can be a veteran with just a few years of usage. This is not me, but I have a few thoughts.


Personal preferences are great for many reasons (too many to list here), however they are not equivalent:

  • Some things are objectively better than others, not for a given person, but for a given work.
    • Some things may:
      • theoretically achieve the same results
      • practically make no difference to the computer
    • This doesn’t mean that the user will have the same productivity with any of them.
  • This point becomes multiple times more important when collaborating with other users of different preferences:
    • Nobody should remove the right of preference. Nevertheless:
    • Nobody can prevent the mess from not using some agreed conventions.

Plugins are great also for many reasons, however they should not alter the conventions of the target application.

  • Overusing a plugin means missing from Logseq’s power.
    • A plugin can enhance the application, but cannot replace it.
    • Before using a plugin, check the alternatives, starting from the application itself.

  • Inline references are internal (or hard) associations.
    • The reason is exactly because they appear in the text.
      • They don’t appear in the text by coincidence.
    • Removing the text of the reference makes the note problematic.
      • The note is associated anyway, no matter if the text is turned to a reference or not.
    • If you express hard associations with tags, you are probably using the wrong application.
  • Tags are external (or soft) associations.
    • The reason is exactly because they don’t appear in the text.
    • The note suffers nothing by not mentioning the tag.
  • The above are not facts from Logseq’s code, but from its design.
  • If a hard association doesn’t appear inline, don’t make it a tag, rephrase the note.
    • References are not only to organize the notes, but also to reconsider them.

Below is a weak choice for a reference, which explains the difficulty in the specific case:

Here are some alternatives, depending on the desired focus:

  • code-learning
  • code-teaching
  • coding lesson
  • coding education

The difficulty to choose the proper reference can be an indication of a problematic note, e.g.:

  • This is not accurate.
  • As a coder myself, I can assure that non-coders can think fine.
    • Some coders are self-taught.
  • If someone doesn’t already know how to think, can never learn how to code.
    • Should learn to think before learning to code.
    • LLMs neither think nor code.
      • They are not taught either, they are just trained to put code together.
        • Nobody needs to learn that.
  • Coding is more about the language than about thinking.
    • Coding in different languages can lead to very different ways of thinking.
  • Coding is only one lesson for thinking. There are others and more important.
    • Thinking for oneself is more important than thinking in a specific way.
4 Likes

I use tags specifically for:

Contexts e.g., #home #work #email
People … #@johnsmith
Statuses… #inbox #review #waiting

My variant on GTD.

Everything else I use [[page]] as I’m usually linking content rather than contextualising.
In Logseq it’s purely visual so it’s never critical to your graph.

In fact, I am also confused, because apart from the style, the functions of the tag and the cited article [[]] seem to be consistent. From a product point of view, this may be over-development?

to me this is actually a power-feature. Like others, I mostly use tags as visual labels to categorize blocks, and use [[wikilinks]] in the middle of a sentence. As they work similarly and only have a visual difference, I just pick one depending on the context (label vs inline) and still find all relevant content when I go the tag page or use a query.

for my own use-cases, I only find benefits in the ‘logseq way’ (tags=links) vs the ‘obsidian way’ (tags and wikilinks are different entities), but since the topic often arise, I wonder :
what is the gain and purpose of having tags and links being different entities ? when would we need to have different outputs from #art vs [[art]] for instance ?
and also are there specific workflows that do not work when tags and links have the same function and lead to the same page as in logseq ?

2 Likes

That’s exactly how I use them too. It’s purely visual, but very helpful. So I can write in a sentence e.g.,

It is appropriate for [[dogs]] to be appropriately trained.

Meanwhile, I can quote an article from the web:

The big bumper website for canines #dogs #training #toreview

Tags for me tend to be ways of classifying information I’ve found, whereas wikilinks are paths to more content related to the information I’m adding. However, it’s great that they’re interchangeable as, ultimately, I want to see everything relating to dogs whether tagged or linked. I never liked Obsidian’s approach; tags always felt second class compared to links and I could have two sets of content relating to #dogs.

If you feel you need to differentiate between tags and links, you could always come up with a way of naming tags differently. For example , prefacing tags with an underscore #_myinfo vs [[myinfo]].

2 Likes

In a sense I do the same as @DavidS

I use tags mainly for visual distinction. And mostly for broader groups or to stand out.
For example when I write a reflection note, I’d tag it with reflection. This is just to group the note and more easily retrieve it later (for example my reflection of the past 7 days)

This is exactly true for my use of tags. It is also not about the note itself, but the group of notes associated by the tag. Like in the example above, to review all my reflections.