Split Vault into "Polished Zettelkasten" and "Messy Work-in-Progress"?

[walt74]

Question to the community: I have an extensive vault featuring >5000 notes of all kinds (from atomic zettels in a zettelkasten to verbatim copies of articles to lists of movies to, uhm, ranked lists of nougat/chocolate creams, to book notes to lists of names for fiction to rough storyideas to MOCs and HUBs and dailynotes, weeklynotes, not to speak of templates and my collection of academic papers and books in PDF). I use this vault to write blog posts and develop idiosyncratic ideas about AI as an interpolative archive, theories on psychology under conditions of the digital era (think AI-psychosis and social media virality and stuff like that) and some writings on classic philosophy, but also commentary on current politics and power dynamics, then i also follow climate change and have some more pragmatic/nonabstract takes on economics (albeit this not being my main thing), and so on and so on and so on.

Now, i’m getting pretty heavily into ChatGPT. I use it quite pragmatically, write up some ideas and outlines in my dailynote and just throw that into the text interface with the prompt to give me thinkers and theories from classic or modern philosophy that correspond to variable degrees with my ideas. The results, combined with my user prompt (“you are a friendly but critical academic” etc) are honestly very usable and, sometimes, actually: crazy good.

All of which makes me rethink my perspective on my obsidian vault. Until now, i always sidelined rigid structure for chaotic rediscovery: My notes are heavily linked, i have MOCs featuring dozens if not hundreds of notes and i just “surf my knowledge graph” to form free associations according to my current ideas. Which is fine, but i also always wanted to “clean up the mess” and at least get it somewhat straight. I think in tandem with my ChatGPT excursions, that time is now.

So, i think about splitting up my vault into a (very) rigid Zettelkasten-like system to develop philosophical/theoretical ideas, frameworks and theories regarding AI, Memetics and SocMed-Psychology, and a wild growing vault? I hesitate because i value even very rough drafting and i never throw anything away, and if i split the vault i would not know where to, for instance, put my dailynotes where i jot down everything from foodie receipes to dates to philosophical theories. Do i keep a journal in my garden and simply cut’n’paste to my Zettelkasten? And why not keep the Zettelkasten integrated but seperated by tags? It’s just that ChatGPT-output would add another layer into my work and frankly, it’s getting a bit overlayered, if you catch my drift.

Now writing about it, what comes to mind is a hybrid solution: Keep dailynotes and all the mess intact, including booknotes and general everything that is work (vault like a workbench), but export everything that is published or publishing-worthy or “zetteled” into a seperate vault, the shiny polished thing able and ready to represent.

Any other ideas? Anyone got the same problem? Anyone already implemented solution for such a problem?

I searched the forum before and there are some entries about splitting a vault, but they all seem to adress something like “work” vs “private”, not “polished” vs “work-in-progress” (under a save-everything paradigm in which you delete nothing, basically).

[vics]

If I’m understanding you right, I struggle with a similar hoarding / prioritization problem. I’m over-organized, but the two changes I’ve made that have worked the best are:

1. Like you said, having a very rigid “Zettelkasten” with alphanumeric IDs that keep my thoughts “in order” as they develop over multiple notes. (Bob Doto’s posts on Folgezettel are great for this.)

2. Using a Johnny Decimal-like “index” to categorize other types of entries in my system.

So, sticking with AI, “Artificial Intelligence” is a header in my Folgezettel (for polished, developed notes about AI), but I also have 22. AI as a category in my Index to organize AI prompts, instructions, conversations, drafts, etc. that are elsewhere in my vault.

The “header” and “category” are both essentially link tags (they each have their own note), so although I use folders, you don’t have to.

I’ve tried splitting my vault a few times, but I reference my polished notes often enough that switching between multiple workspaces was too tedious.

[chriswaterguy]

Why not use folders and/or tags?

A vault is just a folder, after all – but having the polished and messy folders inside a single vault allows for easier discovery and transfer. And links continue to work when you move a file.