Hadn’t seen this here yet, a co-worker of mine sent it my way so I’m just spreading the word. Looks interesting, to say the least! Anyone tried this out or had any other experience with it yet?
You must log in or register to comment.
Seems a neat way to lose everything
I can sort of see the appeal, but its not for me. If anything is ever going to rename files for me its going to be a script that I’ve either written or at least read top to bottom. Not a blackbox inference engine, and especially not one based on an LLM.
Why rename the files when you could just categorise and index them…?
This seems unnecessarily destructive.
I’ve been long looking for this rename feature.
I have so many files that are titled
Document-2022(1).pdf
or
contract(1).pdf
Really wish that they had descriptive titles so I can know what’s in them without having to open them
It would be better to have this as a FUSE filesystem though - you mount it on an empty directory, point the tool at your unorganised data and let it run its indexing and LLM categorisation/labelling, and your files are resurfaced under the mountpoint without any potentially damaging changes to the original data.
The other option would be just generating a bunch of symlinks, but I personally feel a FUSE implementation would be cleaner.
It’s pretty clear that actually renaming the original files based on the output of an LLM is a bad idea though.
