Tagging can be cumbersome and trying to create relevant tags is annoying, I quite like having both vague and specific tags.
AI could read over your file (using an API) and suggest tags that you could use, then by clicking on those tags they are added to your file.
Great idea!
I imagine that AI can also see the correlation with other notes and properly give relevant tags to link to other notes. Better if it could summarize the linked notes.
If AI can find those relationships, it ought to be able to do so in real-time, making it unnecessary to pre-process the notes to find tags to add to the note.
Or, if we really want tags for some reason, why can’t the AI develop its own tag vocabulary instead of being stuck with ours?
Tagging exponentially slows down my notetaking, so an AI that could help me apply my existing tags would be invaluable.
I know what you mean. I’ve struggled with that same thought and here’s how I make everything make sense again:
AI tags provide an objective ‘source of truth’ for your notes, identifying themes and connections across your vault.
Manual tags, on the other hand, offer a personalized, opinionated structure tailored to your specific needs and purposes.
So to build on your point, I can see a world where we CAN do away with hyper-specific SOT tags.
Perhaps a more useful AI Tag plugin would take an additional input param that defines the “style” of tagging you’re aiming to achieve - could be a pre-defined system, or a hard-coded list of tags (essentially a categorization problem), or even a vague idea of HOW you’d like to use the tags.
For example:
“I would like high-level tags that complement my current vault structure. My current workflow includes dev projects, home device maintenance, family finances, and learning”
This plugin has this functionality and a bunch of other AI-powered organizing tools:
I’m trying it out in the self-hosted mode and it seems to work pretty well for this use case.