I’m trying to use org-roam to solve a concrete problem, but I am struggling to understand the most vernacular org-roam way to do this.
What I want conceptually: at work we have a number of systems that have a surprising array of idiosyncrasies. I tend to learn of these idiosyncrasies slowly at random times. What I want to do – originally I thought of doing this in the context of logseq – is simply keep a daily journal, which I really do, and write these things down as I learn them, but when I do so tag them with the relevant system or topic. The later, when I need to review what I know about a given system, and perhaps even formalize and synthesize my understanding I could load up a list of the tags that would be pulled from the journal notes.
In logseq, this can be done by adding a tag, but org-roam doesn’t really have a concept of tags that is meant to be used this way. Tags seem to be attached to a file header. So then I can just do this with links. I.e. I could tag each header in my journal with a link.
I’m curious if anyone is using org-roam in a way similar to this? Any suggestions?
@nobiot: I’m sorry if these are basic questions, but I’m having trouble understanding the role of tags. If I look at the org roam manual and do a search for “tags”, I see tags used in many different contexts, but mostly as it relates to a notion of filetags. E.g. this section:
So from that document, I’m having a hard time understanding how org-roam iteracts with org tags that use the :sometag: syntax. What I’m after in this case is the ability to collect all nodes that are tagged with some tag and view them in one place, something akin to what happens in logseq if I click a tag, i.e. I get a page with a set of “linked references”:
The trick with org-roam-node-find that if I understand correctly would show the tags in the org-roam-node-find search minibuffer is good but not precisely what I’m after here.
@nobiot: I edited my post above to include an image of what I mean from logseq. I think this might be similar to denote’s idea of a dynamic block of notes, though I haven’t played directly with denote yet.
Thanks. Just seen it. What’s the purpose of this for you?
If I want to have a list of nodes for a certain tag, I will just use org-roam-node-find and navigate to one of them. If you want to retain the list, I’d export it to a buffer with Embark.
Here is an example of showing my notes with #reference tag, put into a persistent buffer with Embark (so I can go back to it for further navigation). The window on the right is one of the notes opened from the list.
For the screenshot, yes. Embark is big but this sequence works without extra configuration: I only have this for Embark below. I never use “C-o” (I don’t remember to use it. Now that I am saying it, I might delete it or swap it with embark-export with something like M-e). I would just do embark-act (“M-o”) and press “E” to embark-export.
Just as context, I don’t really use Embark, either, and I don’t rely on tags. I just use my version of Consult-grep (via xref with my own package I name “Ren 連”) on any word or phrase. That’s my “links” and “tags”, really. I have mostly stopped using the link syntax from Org or Markdown.
Here is an example of searching for “bacteria” (case insensitive).
embark-export could turn this into a list, just as an illustration, but I rarely do it. The minibuffer with Consult and its preview, as the image above, does the job for me.
I understand the trouble. In Logseq, I do not see much difference between tags and links [1]. For Org-roam, tags seem to have a different meaning and usage from Logseq.
For your purpose, I would have a note for each system and write my findings there; no links or tags — I don’t write stuff for long-term memory in a daily journal. One is for storage; the other is for flow. If you have a reason to do so, I’d use a normal link and use backlinks to display them later.
Here is my definition and guidelines for tags in any note-taking system, which I apply to Org-roam (as well as any other). I don’t use tags much.
If you’re anything like most people who first come to [[Logseq]], you may be wondering what’s the difference between a [[page link]] and a #tag
Truth be told, the difference is primarily visual. For the most part, they are both the exact same in functionality. They both link to another page. If you click one it will take you to another page and both will appear on the linked references of that page as well. ↩︎
thank you for the hint to use embark-collect. I have been going around the idea of automatically generating lists of nodes in soo many different ways (including the transclusion-fn library
@nobiot thank you for sharing so much. There’s a lot to digest there. May take me a bit of time. From the screenshots you shared, it looks like you are using md and not org format, is that so? Just based on the title, I assumed org-roam required org files.
(I’m also curious what font you are using, and modeline…)
Take all the time you need. Emacs and the communities (including Org-roam) will be likely there.
This is correct. I use my own extension Md-roam. It works and has been for me, but it’s a hack. I won’t support everything Org-roam does – eg. I don’t use the dailies feature so I don’t think Md-roam works for it.
For my configuration, you are welcome to view my “attic” or “shed” (my configuration); not organized for exhibition but should be good enough for browsing.
Mode-line is another rabbit hole and time sink. I have settled to using the built-in with a handful of changes. For the changes, I have decided to rely on the following packages (moved away from my own):
