I recently took a new job. The employer is a Microsoft shop and would prefer that we use OneDrive to store work files.
I vaguely recall reading a post that recommends against using OneDrive for a DevonThink database—but I did some Googling just now and it seems that my memory is incorrect—it seems that OneDrive may be supported?
For now, I just want to create a database, store that database on the OneDrive folder on my Mac, and use it on one machine. For now, I will not sync that database to iCloud or any other cloud service. Will that be a problem?
In the future, I may want to index folders on OneDrive.
And I may also want to sync to another Mac, iPads and/or an iPhone.
Will any of that be problematic?
I suspect if I decide to sync between machines, I’d be better off doing Bonjour syncing than using OneDrive.
For now, I just want to create a database, store that database on the OneDrive folder on my Mac, and use it on one machine. For now, I will not sync that database to iCloud or any other cloud service. Will that be a problem?
Nothing has changed in cloud in this regard. You should never put DEVONthink databases in any cloud-synced location or you could irreparably damage them. To this end, DEVONthink disallows creating, opening, or syncing a database in such a location and issues a warning about it.
In the future, I may want to index folders on OneDrive.
If maintaining tags on the documents isn’t important, then yes, it’s feasible. See…
I suspect if I decide to sync between machines, I’d be better off doing Bonjour syncing than using OneDrive.
You couldn’t sync to DEVONthink To Go via OneDrive in any situation.
Yes, you technically can index OneDrive items. However, they should ideally be stored locally, not just in OneDrive on the cloud.
DEVONthink tags are automatically applied as Finder tags on indexed items. If OneDrive strips them off, then yes they would disappear on the documents in DEVONthink.
When I used to index it was with OneDrive. I always applied tags locally and never lost them. But as @BLUEFROG says do NOT store your database there. It will be a disaster.
I think I moved from indexing maybe a year ago. Tagged files in DT never got stripped of whatever I applied to them. Again, tagged locally in DT. Putting DT aside I VERY rarely assign tags in Finder. It’s probably been at least 5 years since I last did.
Regarding the disappearance of tags on indexed OneDrive folders: what about documents assigned to tag groups or replicated to a group within a folder?
I like to keep a running chronology of docs for a case in a master folder with sub folders holding replicants of some of those chronology docs for categorization.
If I were to move this data to OneDrive and index, would the replicants in groups in DT3 survive the OneDrive sync?
I get that this is a OneDrive issue. I also understand that OneDrive in the finder will not reflect replicants in DT3. but I was also under the impression that replicants in groups and tagged documents were functionally similar - does OneDrive jeopardize replicants stored in groups?
Hi I am using locally synched OneDrive folders that I have indexed to my, equally local DT3 database. Everything works very well except for one thing: when I check for file integrity via the DT3 command, about 50% of the files in my OD folders show a “Invalid file integrity checksum is different”. Should I worry, since these files do not appear to be corrupt?
I just wanted to chip in as I have a similar OneDrive environment with my employer. I also found that I could index the OneDrive synced drive, but wasn’t able to store Mac-specific metadata such as labels and tags. Also, OneDrive doesn’t always download everything locally, and does this on-demand behind the scenes using Apple’s File Provider framework.
My workaround was to use Chronosync and have that bidirectionally sync/clone my OneDrive volume. I point DEVONthink (via indexing) to that copy. That way, I can use tags, labels, etc. And everything really is available locally for offline working.
Chronosync offers quite powerful rules for excluding files from the sync (to save disk space). This also means I can create groups with notes, AI-generated DEVONthink 4 summaries, etc. within the OneDrive directories, but have them private and never being copied across to the actual OneDrive.
I get around this a little bit by making a very specific smart group(e.g. name contains …, extension is xlsx, item is indexed) in DT for files that are in the one giant OneDrive folder that I index. It’s very handy to put those smart groups in other groups with similar topics. It’s not tags but it does some of the work of tags.
Another unexpected benefit from this very tentative foray into indexing parts of the corporate Sharepoint is seeing, in a convenient place, what files have been changed recently.
One thing I like about your description of Chronosync is the ability to exclude files. When I indexed the main folder for my department I had to somewhat painstakingly exclude the folder of hundreds of photos and videos. I was trying to be extremely careful as I am not a confident indexed file user and I was wary of making some tired mistake and disappearing all the files off the OneDrive and suddenly having them imported into my DB.
Yes! I remember having to do a lot of trial and error when I originally set things up… Chronosync is particularly nice as it’s quite easy to turn folders on and off, as well as setting rules (such as “don’t copy audio files” and so on.
Yes - I totally relate to the worry of accidentally deleting files. Worst of all is moving them. With deleted files you can restore them in bulk from the OneDrive web interface; but if something accidentally moves them then that can really be hard to trace.
I found early experiments with Chronosync were tricky because OneDrive was a bit buggy and sometimes files didn’t show locally - which could result in them being deleted on both sides. Chronosync now supports OneDrive more natively (e.g. it understands the idiosyncracies of the OneDrive sync folder); I’ve not had any issues for the last couple of years…
Because I am working in a gig to gig situation, I don’t have to manage the indexing issues across multiple years, new computers, or system rebuilds. When a gig ends, I copy the indexed files into the DT database, do some basic tagging and organizing, back up the whole project OneDrive to an LTO, turn off the lights, leave the key at the front desk, and move on. I usually have the last two jobs’ DT databases open to raid for useful docs and smart folders.