#13

I do what @xavierroy says as well, excluding the folder with -path:foldername. Sometimes I also use -file:part_of_filename if I want to exclude a group of notes that have a common piece in their name.

1 Like
#14

I have reservations about strategies that exclude paths like that. It doesn’t seem that a process requiring exclusion of various paths is scalable. It can lead to quite complicated and increasingly brittle queries as the number of paths to be excluded increases over time, multiplied by the number of queries that must exclude them.

I already experienced this very quickly with tracking tagged tasks for individual people and projects using embedded queries in their notes. Every time I added one note that needed to be excluded it required an update of N notes containing the query, where N was becoming increasingly larger. And since the queries were almost identical except for the person-/project-unique identifier they had to be updated manually, which became tedious fast.

It may work for now in the small scale but I’m not sure this is a great long term strategy.

1 Like
#15

It would be cool if we could have variables in our notes. Like, if I wanted to show someone my notes I would need to exclude all of my clients. My embedded searches would exclude the clients variable, which I just update as needed.

I agree that in general those sorts of ritualistic exclusions generally just make it so you need to type more and more though. This is also why I’m preferring atomic pages to using nested headers and the section/subsection search. I don’t want to have too many different rules governing my notes that would produce false positives in one case and false negatives in another.

2 Likes
#16

yes, you have a point. Which is why next to excluding paths I also use excluding files with common name patterns. w.r.t. @Daveb08 original question, I use [searchterm] -file:Daylog as it excludes all my day logs, regardless of whether they ended up in the same folder. For other types of notes I use -folder as naming conventions can get in the way of things as well.

1 Like
#17

One way you can go about this is using templates that include yaml metadata of your interest (e.g. clients). You can then filter using the dataview plugin. Not so much more typing and in return you can get neat embedded queries :grinning_face_with_smiling_eyes:

3 Likes
#18

Thank you for explaining your process. Do you know if that method is outlined ANYWHERE in the Help documents. I couldn’t find it. Or did you gain the knowledge elsewhere? Again, thank you for taking the time to respond.

1 Like
#19

In the Obsidian Help Vault, the page [[Search]] describes the various ways one can use search, including how to embed it in a note. Search is a core plugin, you can see in your Obsidian settings if it is enabled in your Obsidian instance.

1 Like
#20

I know, but nowhere did it mention ‘-Folder’ as an option. I assumed that was something you picked up elsewhere than in Obsidian Help. I saw the section, but not the option. I just thought the section could have been a little more expansive. Certainly NOT blaming the Devs, I think they are amazing.
Thanks again.

1 Like
#21

I probably picked it up browsing the forum myself, but the Obsidian Help Vault mentions path, file and tag etc under Search operators, and the use of - to negate things and using regex in the section above it.

1 Like
#22

This is also why I’m preferring atomic pages to using nested headers and the section/subsection search.

I’ve also found this to be very beneficial. It reduces coupling between notes, making it easier to make changes in notes without breaking links/queries/etc in other notes.

Just like software packages, notes should have high internal cohesion (the note should be focused on one core idea, not necessarily short but can be) and low coupling between notes.

Coincidentally this is a fundamental principle in the zettelkasten method.

2 Likes
#23

How do you group atomic notes into a larger structure?

I’m still very new at Obsidian. Currently, the two pillars of my workflow are Daily Notes, and a Dashboard. The Dashboard attempts to be a ToCo of everything in the vault.

So let’s say I break the Dashboard into many, many atomic notes. How do I do that and still retain the overall structure? Keep the Dashboard, with that document consisting of nothing other than links to individual, atomic notes?

1 Like
#24

This thinking only goes as far as making pages amenable to searches and reducing false positives. This doesn’t always matter. I think that’s especially true in project pages vs. knowledge pages, or for how much of a “verb” a page is in your workflow vs. a “noun”

I like having contextual dashboard pages for my projects. For example, one of my projects has a few embedded searches on it to bring in the relevant notes from elsewhere. I wrote this post a while ago but it actually applies quite nicely to Obsidian as well if you can structure your searches well. https://www.roambrain.com/brainstorming-with-myself-systematic-creativity-in-roam/

Alternatively, a contextual dashboard with commentary. Instead of putting everything on one page, it links to all of the smaller pages. You can even ![[embed]] those pages if you want them all viewable from one location but not muddying up search results. The embedded search approach is more a way to “automate” that.

3 Likes
#25

You can maybe even call it functional note taking. I see a lot of parallels between this modularized note taking approach and functional programming.

2 Likes
#26

Good insight. I’m not sure how applicable to my own work, but useful nonetheless.

Your description of your work could apply to mine as well: Juggling many projects simultaneously, with information coming in from a many sources. I’m in marketing, rather than product management, however.

I’m thinking of scaling via hierarchical dashboard pages. One dashboard – and folder – per project, with a dashboard-of-dashboards to manage it all.

1 Like
#27

I’m experimenting with a template vault from @mdee that provides dashboard views for projects and tasks using the dataviewer plugin. It’s got a lot of promise but there are some issues where the query syntax in the plugin may have changed a bit recently (unclear) and there may also be a new bug where the views don’t automatically update (also unclear). The workaround for the non-updating views is simply to close and re-open the note showing the dashboard.

1 Like
#28

Funny… I arrived here looking for an Obsidian+Figma workflow and ended up learning a ton as an Obsidian newbie.

Still working on my strategy of balancing atomic notes vs dailies, so thank you.

1 Like
#29

I’m glad it was helpful :slight_smile: I use both Figma and Obsidian as well… you can actually embed with iframe! Just copy the embed link from Figma and then you can make a page for quick access in Obsidian.

2 Likes
#30

Yes! Thank you for this. This makes all the sense in the world to me.

1 Like
#31

Another expat here.

I didn’t know you had switched over. I’m following behind you. About 3 weeks ago I started working on learning Obsidian. I’m all in now but finding some of my work practices/mental models to still be trying to follow Roam rather than shape to Obsidian. I appreciated your conversation here as helping me further develop my practice (I was a very heavy DNP user as well).

4 Likes
#32

RobertHaisfield, I found your compare/contrast piece very enlightening with respect to how Roam and Obsidian differ in their approach to handling notes. In light of your criticism(s) of Roam, can you share with us why you didn’t switch to Logseq instead of Obsidian. It seems to me that Logseq does not have the constraints that you mention that Roam has, yet the functionality of Logseq is near clone-like to that of Roam.

1 Like