For example, I noticed that in the demo access app, there's a note about cooking, and it has 4 tags: - `baking` - `cupcakes` - `oven cooking` - `recipe`
This would get out of hand quickly.
There should be a hierarchy of tags (categories): `cupcakes` in `baking` in `oven cooking` in `recipe`
The only tag needed in this case for the note would be `cupcakes`
I own and operate a "list-taking" app[0] in which every list/kanban-item can itself be a list/kanban.
I currently use it for things I'm the creator of -- tasks, story outlines, etc, but looking to introduce 3rd party content for task management (I want to see GitHub tasks from work next to my own tasks) and, as you say, knowledge management of things like recipes or music.
Items could be part of one or multiple hierarchies. A list of "cake" recipes could be under both "baking" and "party essentials", and music playlists could include other playlists.
As you can tell, this can become convoluted in my mind, and so if that's something that's interesting to you (or anyone reading this), please reach out and let's discuss! hn at nestful.app
"Spontaneous productivity" mirrors some of my own thinking on the subject, especially the JIT and bubbling aspects and how they work together. I haven't seen how it works in the case of Nestful, but I'm keen to try it out. It may adjust the design principles guiding development.
IMO it would be interesting to try to combine the two approaches (curation + auto tagging).
It starts out with the user scaffolding an initial hierarchy, then (after enough usage to provide meaningful data for ML predictions) the ML model predicts on subsequent entries, and asks the user for approval (which feeds a reinforcement learning model)
Hierarchies get out of hand quickly too. You will soon find that different people (or the same person at different times) create different hierarchies for the same thing and that the same thing belongs in multiple places.
I have been experimenting with different representations of data in Neo4j, Markdown, and Orgmode. I even tried cludging the polyhierarchies into different file systems using symlinks and tagging,
I'm still researching for better storage techniques.
I want a good mix between hand editing, but robust machine readable formats. Orgmode works pretty good, but it's fairly complicated to parse, and I think it could be improved.
The retrieval and search part could be improved with RAG, but I don't have the hardware or time at the moment to hacking around with the compute intensive AI stuff.