I believe I know the answer, but I wanted to confirm and ask for additional ideas.
I understand the position that the team at DT has taken regarding not supporting the myriad of syncing and cloud storage solutions out there. I also understand that there are 3rd party options to provide WebDAV feature sets for cloud storage solutions that don’t inherently provide it. For example, OneDrive is not supported, but the way to do so is via WebDAV. I will add that I’m not sure I completely understand the purpose of the WebDAV function, so there is a knowledge gap here I need to do more research on.
I’m trying to solve the question of scaling and large datasets. If my combined dataset is larger than my hard drive capacity, are there any recommended strategies to still have access to the meta-data without needing the data to be “online?” For example, I index my files, relying on “external storage” rather than saving the files within the DT database structure. Is the meta-data from the index sufficient to execute queries without needing the files to be “online?”
One possible use case is to initially download select folders from OneDrive, perform an index, and then change it to on-demand sync, so the files are not consuming space.
Where is your OneDrive folder located?
You shouldn’t index files that aren’t local or on a connected external drive or NAS.
And yes, the metadata and text of indexed files is still searchable. However, if the files are not accessible to DEVONthink, you will only see a small thumbnail and a File Missing notice with the last known file path. If they’re on a connected volume, you should see a Mount Volume button as well.
One possible use case is to initially download select folders from OneDrive, perform an index, and then change it to on-demand sync, so the files are not consuming space.
I’m not sure what you’re referring to here as DEVONthink doesn’t support shallow syncing. Are you referrig to DEVONthink To Go?
My OneDrive folder is located on my local hard drive. I’m not currently in a situation in which I need to be worried about the size of the dataset, so my inquiry is to proactively solve for a future problem.
Are there any recommended strategies when it comes to maintaining large datasets?