This question is probably more specifically for @BLUEFROG, since I’m referring to your opus magnum, the manual. Of course, I’m also happy if anyone else has the answer! In the manual, I read this, and it made me think:
INDEXING AND SYNC
Often people index content from the local
repository of a cloud service like Dropbox.
The question then arises: “If I have my data
on Dropbox’ servers and I sync to Dropbox
via DEVONthink, aren’t I just using twice
the space?” The answer is practically, yes.
I’m rather new to DT, and thought I had understood the concept of indexing, but I still want to double-check: what is synced to whatever remote location I chose is the index, right, not every single file? I.e. if I index a text file, I’ll probably use twice the space. But is that also true if I index an image file?
Here’s my usecase: I’d like to create a database that indexes the files I have in my iCloud folder. Some of those are huge, so I don’t want to have them uploaded twice, but I’d like to be able to search for them efficiently from my phone. I thought that when indexing them on my Mac, DT would create a database and a searchable index, and sync that to the iCloud. I could then sync that same information to my phone, and search on my phone. And when I decide to open a file that DT has found, either it will already be on my phone, or it will be loaded from iCloud on the fly. Is that right?
DEVONthink syncs the content of indexed documents by default, regardless if they are indexed in a cloud-synced location or not.
Does that mean it’s technically double the storage? No.
Does it mean that’s practically double the storage? Yes.
As is noted, there is an option in the sync location’s settings to disable syncing the contents of indexed items. However, this options should only be used when syncing between two Macs using the same account on the cloud service. It should never be used if DEVONthink To Go is involved as (1) it has no way to access the files outside itself so (2) there would be no contents to download into its database.
Thinking about it over the weekend, I wondered if a “dirty” solution might work: create a database on the Mac, sync it to iCloud without the content, and then sync it to iOS. Would searching work on the iPhone? If I have the filename / folder, I could manually go and get the right file, but at least I would have a good search function. I understand it’s not how it’s supposed to be used, but would it work nonetheless? If not, a simple “no” as an answer is enough, since I’m already aware that this would be a non-standard usage.
A shallow sync does allow for documents to be found in a search.
I could manually go and get the right file,
Do you mean you couldn’t, because that would true – you couldn’t.
In this configuration, DEVONthink To Go would only serve as a pointer to items accessible in your databases on the Mac, i.e., “DEVONthink To Go says I have this file when I search. However, it’s only available on my Mac.”
