update: The below is modified from the previous version, which used an alphanumeric folder-naming convention. This was too much maintenance. Removal of the sub-folders was done by adherence to a file-naming convention, whereby whether a file is my own creation or that of another is identified simply by the last name used in the file name. This allows for, in most cases, a single folder level.
I like experimenting with different folder hierarchies and I share here my newest creation. Basically there are two levels:
type
type_name_YYYY-MM-DD_title
The motivation behind this structure is that I absorb information from a wide diversity of sources and create notes and content in correspondingly diverse ways. This folder structure lets me use a single vault with the folders acting as “mini pseudo vaults.” See a model below for a more concrete example:
?questions
@ramblings
academic
classes
letters
news
personal
journal
goals
poetry
professional
references
lists-routines
organizations
people
plans
recipes
shows
speeches
stories
visual-art
You can see that each “type” or “space” receives a number which has been assigned solely based on alphabetical order, but you could use whatever numbering system you like. The second level folders are then qualified with a letter corresponding to its sub-type, i.e., source, note, or creation (creation = output/project); in some cases “creation” is replaced with the specific type of output I am interested in generating in that space.
The next obvious step is that my filenames mirror this convention so that when (if) I create links between notes, I can see at a glance what kind of a note I am linking and therefore what type of information I’ve included in my draft as its foundational content.
Anyways, just a fun thought experiment. Hope it’s useful for somebody!
I LOVE organizing data into folders, it tickles my OCD in just right way. For what I’m doing in Obsidian, I’m finding more and more that folders don’t work for me since my data often fits into multiple folders (eg Where would a letter about poetry classes fit?). Tags would seem like the solution to this, since each piece of data can exist within multiple tags, but then you need to organize the tags, which has the same problems as organizing the data. What I’ve come around to is using Obsidians most powerful feature: linking. Instead of creating a folder called “Academic” I would create a note called “Academic” with a Dataview block that shows any linked files. Any “Academic” notes would then have a link to it, so it gets added to the list of links on the Academic note. When I’m looking for an Academic article, I’d use search to bring up the Academic note, and then click the link in the list of linked articles. When the list gets to big, time to create sub-categories using the same process. Though I still use folders for types of notes that comfortably fit within only one category (and I can then use a custom template for all those notes), most of my notes are now lumped together and I find them through search and clicking links. It’s new to me, but has been working very well. Something to consider exploring, perhaps something in between this and what you have ends up being your ideal situation.
Lol, this got me pretty good. I feel compelled to answer this question. I guess I would start with what it is: it is a class, so that’s the folder in which I would put it. In my system, this would work because my folder about poetry is strictly used for poetry I have read or poetry I write. Classes go in classes. If I took useful notes about poetry from that class on poetry, I would add them as files in the 2_classes > 2b_notes folder. If I wanted to link to them from some notes in my 7_poetry > ... folder(s) then I would do so.
Now that’s out of the way, let me thank you for your comment and interest. Certainly linking is a powerful feature. This folder hierarchy and file naming convention helps me move toward making better use of links which, until very recently, I found significantly overwhelming to the degree of not using Obsidian or even reading material anymore. But I appreciate your share. It helps me in my continuous journey to better understand links and how I could use them and of course I can see how they are useful and your flow seems to have a lot of value and simplifying potential.
Right now, in order to facilitate use of Obsidian, I am using folders as my primary organizational tool and line:() search combined with path: search, links extremely carefully and selectively, and tags very specifically. For now, not using Properties.
Makes sense, I use Obsidian in a much different way today than I did early in my adoption of it, and I suspect a year from now I’ll have found even better ways of meeting my needs with it. If I’ve learned one thing from my Obsidian journey so far it’s to do things in a way that’s modifiable, so I can pivot in the future should the need/desire appear. Creating tags or a linked TOC/MOC files from a folder is relatively easy, so organizing things in folders isn’t wasting time if you change methods down the road since it will provide a means to the next option. It’s a great tool to grow into!
Your post was useful and interesting, thanks. I’m kind of hitting this pain point in my own management - I started with erazlogo’s folder structure (I wish I could send erazlogo a cake or something for the whole website on managing research) and it aided my understanding and adoption of obsidian. But then y’know some parts I didn’t need, so I trimmed down the folder structure. And while I initially had Zotero inputs only, I started adding Readwise inputs, YouTube inputs, etc.
And now I have a Research vault that looks like:
01_sources
**Articles
**Books
**Images
**Literature (notes import from zotero integration)
**Video
a bunch of random notes in the root folder
This is part of the folder-hoarding problem that led me to obsidian in the first place, because now I’m getting confused about what types of files go where. If there’s an html file on the web that has a journal article, does that go to Readwise or Zotero?
It matters because then the file names will be different upon import. And so now I have a bunch of files with different naming conventions; books from Kindle (via Readwise) are Sentence Case Titles but books from Zotero are lastname_title.
I just feel like having a big messy inbox and then processing individual things out to the folders above. But I like the Zotero import to a specific folder with a specific filename.
So I’m curious how are you planning to deal with filenames? Are you trying to set up a taxonomy in the folders that’s going to cover any different type of note? I caught myself setting up a “reference” folder too, and I kind of like your section 8 for keeping track of stuff like Cs, AWP, SHARP, NCTE, other orgs I follow that have conferences that I forget to keep track of.
Thanks for your reply! I haven’t heard of erazlogo’s folder structure. Can you share that? Sounds interesting and I appreciate learning about your own system!
for me these tools serve separate purposes, however, they are both intermediary steps before Obsidian. I don’t use Readwise so I make no comment here, but Zotero is, for me simply something I use to read and archive PDFs for journal articles, and the notes on these go to Obsidian. It is not an issue that the file names may be different in Zotero and Obsidian, but there are ways to resolve this (and I think using fewer tools is one good way).
File names start with the same alphanumeric code as the folder in which they belong. For example: 1_academic > 1a_sources > 1a_authorlastname_YYYY_hyphenated-title. This is the same (as much as possible) for all file names in each folder. Sometimes date or title may not be relevent.
But short answer is yes, there is a naming convention which is as universal as possible, but adaptable for each folder/thinking space.