Help with DT/Zotero workflow for newbie

Hi all. I have searched the forum for insight about this topic but haven’t found a clear answer and I’m really new to Devonthink and thus a little intimidated by all the lingo.

Here is my goal: I want to use Devonthink for storing and organizing all my sources (I’m a historian), especially PDFs of archival documents. I want to be able to use Zotero for citations for these individual archival documents. Should I enter metadata in Devonthink first? Is there a way to transfer that data to Zotero? Or should I use Zotero and transfer to DT?

Thank you so much for any advice you can provide.

sj

Welcome @sfaj
I don’t have a definitive answer (not being a Zotero user) but there are discussions on these forums…

https://discourse.devontechnologies.com/search?q=zotero

I think it is covered in the forum, but you might want to take a look at:

1 Like

Thanks. They recommend Bookends and I’m already deeply invested in Zotero.

Also, the ladies shuttered this resource a few years ago now.

1 Like

Which ladies?

Avigail and Ada who founded and ran DEVONthink For Historians.

Hi @sfaj . I study ancient history and linguistics and use both DT and Zotero. I struggled to find a good workflow and finally settled on the following:

  1. Store all files in Zotero.
  2. Index files with DT.

Store all files in Zotero.
I chose to store all my files in Zotero for a couple of reasons. For my purposes, it did not make sense for me to keep the bibliographic info in Zotero and the PDFs in DT (which I did for quite some time to avoid paying the storage fees for Zotero). Zotero does a pretty good job of getting citation info when dropping in PDFs so that swayed me. But mainly I store my files in Zotero because the PDF isn’t modified when I mark it up in Zotero, but in DT it is. I was also an archivist by trade, so I like to preserve an original copy; this decision allowed me to keep a single copy rather than two. (And of course, I have backups.) Annotating PDFs in Zotero on my iPad is also a nice experience. Eventually, it also turned out to be even more beneficial because I now collaborate with other Zotero users, and I can simply drag and drop the Zotero entry to a shared Zotero library with the citation and the attached PDF to share.

Index files with DT.
Please use caution on the next step if you choose to go the same route: Indexing. What this means is that DT indexes the files or folders you choose (it does not make a copy in your DT database). But the integration is really tight with the Finder in that if you delete the file in DT, it will also do so in the Finder (if you are not careful, see below). So you can probably imagine the potential problems of deleting a file or folder in DT, only to find out your Zotero data is missing.

So with all that said, here is what I do: I index all my PDFs from the Zotero folder structure, but I index each file separately—NOT, I repeat, NOT—the folders. This allows one single advantage: It allows me to move the file anywhere in my DT database without moving it in the Finder. This makes possible the ability to store the indexed PDFs in the folder hierarchy I want in DT, rather than Zotero’s File structure and randomly generated folder names. If you index folders in the Finder with DT and then move a file within the indexed folder structure in DT, the changes will also be reflected in the Finder. But this does not apply to individually indexed files. (A major change on indexing from a couple of versions ago.) So moving individually indexed files in DT does not move them out of the folder structure for Zotero in the Finder. This does make my workflow a bit tedious, but it works for me.

One more thing to mention: If you delete an indexed file or folder in DT, the file or folder is moved to the trash in DT. I don’t empty my trash often (I am an archivist!), but when you do, you will get this popup:

Be sure to use “Only in database” if you don’t want any indexed items in the Finder to be deleted because if DT deletes the '“Files & Folders” of any indexed items, they will be deleted from the Finder as well. Take note of this, especially if you experiment with indexing with only a handful of files.

Well, that was long-winded, but I hope to spare you from the grief of inadvertently deleting any files and folders down the line. I hope this helps you to evaluate how best to integrate DT and Zotero for your own purposes.

Darryl

3 Likes

Thank you! I really appreciate you spending the time to offer this guidance.

I have a different solution to that outlined by @darrylmy . I am a literary historian with plenty of archival work (many images from manuscript sources), and a large library of PDFs of primary and secondary lit.

For me, Zotero hold only the metadata of the publications I work with. I don’t store any PDFs there, don’t annotate there, and so on. That frees me up to use what’s best about Zotero: citation management. I am probably quite a light user of a subset of the app’s functions (my database has about 1500 items), but the app absolutely shines when it comes to preparing first drafts for publication in Microsoft Word. I don’t work in Word until I absolutely have to (note-taking happens in markdown in Obsidian and DT; long-form drafting happens in Scrivener where I just use place holder citations: e.g. {Sfjaj 2025, 123-45}). Only when I export a finished piece to Word will I start linking to the Zotero via the plugin.

All my research materials live in DEVONthink 4 on the Mac, and selections are synced via Bonjour to my DTTG on my iPad. All my reading, note taking, highlighting, annotating, happens in the DT environment and this experience is absolutely unmatched. That way, I get the best in class of what both applications have to offer (in DT, especially search, and smart groups), without having the worry about managing citations in DT (suboptimal) or storing large amounts of data in Zotero (suboptimal).

You can spend a lot of time making the two apps talk to each other in more sophisticated ways, but frankly (for me) that wasn’t worth the effort.

1 Like

@sfaj , my workflow looks a lot like @darrylmy ‘s workflow, except for where I store my PDFs in Zotero. I use a Zotero plugin called ZotMoov that allows me to save all of the PDFs in a folder outside of Zotero (I save them into groups that match my folder structure in Zotero). I think index that group into Devonthink. It doesn’t allow me to move files around the way that @darrylmy ‘s approach does, but anything I add to Zotero automatically shows up in Devonthink, which is nice.