Advice/best practice for setting up DT (for a new situation)

Due to pandemic and other things that seem to happen I’m looking for some advice for best practices how to set up DT. So far I’ve mostly used DT with a number of databases on two Macs that I’ve synced using iCloud (previously Dropbox), I’ve also synced it to iOS but I haven’t really used that part.

I’ve also just dropped snippets, text files and PDFs storing them inside DT. But things might change in the future and I’m looking for some advice how to handle those situations.

Hardware: I will still have two macOS devices but I might need to use different iCloud accounts (work and private), I will also have my iOS devices (connected to my private account)

Services: No Dropbox. Might be able to set up one of my Synologies as WebDav server.

Content: as before text files, PDFs, etc. But I also have a bibtex database (I’m using BibDesk to manage it) and a folder with a few hundred PDFs that would be nice to be able to search/analyze in DT - all of this is a git repo. Also I have several git repos with course material that I’ve written (markdown files, images, some Keynote files, PDFs, OmniGraffle files, iThought files, OmniOutliner files, PNGs, JPEGs) that I would also like to access/search from within DT.

Since I’ve only used the basic side of DT I would appreciate some advice from more experienced people (or pointers to where I can find info about this) on how to handle this situation.

My guess is that indexing folders would handle the repos and PDFs but I don’t really know the best way to store these files so that I can keep them in sync. Dropbox is out of the question, Synology Drive might be an alternative here.

My primary goal is to be able to use this on both Macs, it would be nice if some stuff where accessible on iOS also but macOS is the priority.

I’m open for all kind of suggestions (I realize that using an external disk is an alternative but knowing myself I think it would be better with another solution :stuck_out_tongue_winking_eye:)

Hello. Welcome to Devonthink? Actually, it sounds like you are pretty familiar with your options already!

One option for using DT on both Macs is to store your files in Dropbox (“files,” not Devonthink’s “database”), which I usually put in a designated folder in my Documents location. I “index” all of those files. This has a couple of benefits: files get backed up in Dropbox and if Devonthink is ever out of sync with iOS (often the case), the files are immediately accessible in Dropbox.

“Indexing,” however, requires a bit more thought to carry out well, so I recommend reading threads here and experimenting before you try it out with any mission-critical stuff.

If you are against using Dropbox, as you suggested, then that external drive is super helpful. It can also be indexed, encrypted, and so forth.

Regarding indexing, I suggest you read this: Help > Documentation > In & Out > Importing & Indexing > Indexing and the filesystem.

Thanks for your answers. I’m will need to think about how to set up this, but I have one more question: assume that I have the same absolute path to a folder on both machines and that this folder is on a “cloud service” (in my case it would be Synology Drive instead of Dropbox), is there any advantage/disadvantage of using a “cloud disk” compared to using an external disk? I mean from the perspective of how well DT works.

Would it be better to have “external files” on the disk and sync the database? Would it be even better to have both external files and a DT database on the external disk but still use it on both computers, no syncing at all for that database … just “sneaker net”? Things I didn’t really understand by reading the manual.

Indexing folders in a cloud location may require manual updating of the indexed items in DEVONthink.

Having the indexed folder and database on an external, portable drive and sneakernetting it around is certainly a viable (and private) option. The only two things I’d say are:

  • Use a proper hard drive, i.e., not a thumb drive
  • Make sure you include this drive in your backup routine.

Thanks. And yes, I would use a SSD disk and it would be backed up to my Synology which is in turn backed up to another hard disk (me being paranoid? :stuck_out_tongue_winking_eye:) Not to mention that most of what would be on that disk would be a part of one of many git repos.

Don’t forget to include offsite backups :slight_smile:

Already handled with a second Synology at a different location :stuck_out_tongue_winking_eye: (I have a few files that I really don’t want to lose - not related to DT)

1 Like

Do you use indexed git repos in DEVONthink? I tried it yesterday and there were strange things going on, e.g. after committing I later found multiple records where it should have been only one.

Not yet, I’m still planning for how to set up this but I will try it later (I want to index a few git repon)

Same here. I want to investigate what was going on with the repo later, so far the only possible explanation I could think of is that the .git directory possibly has to be outside of DEVONthink, separated from the working tree. Maybe @BLUEFROG or @cgrunenberg know more about that?

As DEVONthink doesn’t index hidden files/folders (like the .git repo) the git data shouldn’t be touched by DEVONthink.

Is the option to create duplicates in case of conflicts (see Preferences > Sync) enabled on any computer/device? In which folder is the repo located?

No duplicates in case of conflicts. Folder is /Users/User/Projekte/

A Tinderbox file in one repo ended up with 7 records in DEVONthink, but there was more going on.

It took several attempts till a repo was successfully indexed. I tried everything, removing it and indexing again, also with new repos, but it took quite a while until all folders were successfully indexed. Never saw something like this before, a subfolder showed up, but its subfolders and files didn’t. This changed from folder to folder, one time folder A was fine but B not, and vice versa. Verify and repair didn’t help neither.

However I hope I can reproduce this as all folders and files were exported from TheBrain, an app I never used before. Also never indexed a repo before (except a test with a markdown folder which worked well). Weird …

Did you check the sync preferences as suggested? Are the duplicates created after a sync or even after disabling automatic/scheduled sync?

Yes, of course :slight_smile:

Definitely not as I didn’t use my phone.

I can’t prove it but it seems obvious that committing has something to do with it. Unfortunately I didn’t compare the records.

Do duplicates reference the same file?

If they have the same path? Don’t know, I deleted everything. But I’m trying to reproduce.

Yes, whether the paths are identical.

Ah, I did open an old Tinderbox file in which I did a bit too much coding experiments which made it very unstable. So Tinderbox crashed several times after launch - not sure but I think this happened while the Tinderbox in the repo was open which would mean it was automatically reopened on launch and then crashed again. Could macOS’s versioning possibly have caused duplicated records?

The versions are not stored in the same directory, therefore this shouldn’t be an issue.