Zotero best practices

Hey, I just tried out this plugin and I’m loving it so far. Enabling a two-way integration between Obsidian and Zotero is fantastic. Great work!

1 Like

glad to hear :slight_smile: thanks for reporting back

I love this workflow, especially since it appears it can work with most any PDF annotation as well as Zotero’s own beta reader/annotator. However, my ideal workflow would be being able to add annotations to a Zotero PDF (via beta reader and/or from pre-existing, imported annotations), then have all those annotations automatically saved into a destination annotations markdown file (one per PDF, perhaps with an author-date title) within an ‘annotations/’ directory in my Obsidian vault, quotes and pdf page links and all. I could then link to or transclude the MD annotations from any other notes in my vault. If I subsequently added new annotations to the PDF (or removed old ones), those annotations would be automatically added (or removed) to that same annoations MD file in my vault, without breaking any Obsidian links to other annotations in that file. I don’t want to have to think about exporting my annotations or moving them over - I just want to highlight something in a PDF, then know that I can find that highlight in the corresponding MD annotation note.

The Annotator plugin already allows some of this idea (GitHub - elias-sundqvist/obsidian-annotator: A plugin for reading and annotating PDFs and EPUBs in obsidian.) but it’s still in earlier development and having Zotero integration would be amazing.

Does this ideal sound plausible, or even interesting to anyone else? I am not sure if the mdnotes plugin could even do all this or if Zotero itself would need changes. EPUB annotation too would be even better, but I know that ain’t happening :rofl:


Check the bibnotes plugin :slight_smile:

It does not import annotations automatically but it does simplify the process a bit IMO.
You can also update annotations when you make new ones, although there are still a few bugs left.

e.g., updating annotations (syncing with zotero) works but the newly added annotations are added with a missing return line; so you have to add that manually. If you do not, subsequent updates won’t be possible.

You could also comment on this feature request on the zotero forum : [feature-request] Update extracted highlights - Zotero Forums

**I do hope somebody will come up with what you are hoping for :smiley: **



First, I apologize for taking so long to reply. I was not monitoring this forum, but I expected to be notified by email of replies like yours. Either I missed the email, or it never came.

Second, I’m only now rolling up my sleeves and learning to use Obsidian in depth. So take what I say with a grain of salt. Nonetheless, my answer to your question is an emphatic NO!

tl;dr Don’t store files with generic file types in application-specific locations unless the application organizes and maintains these files in an intuitive, transparent way for general access by any other application that might be used to work with them.

Here’s why. Certain file formats are designed to be “open,” which means various different programs can use the same file format. The *.md files Obsidian uses are an example. Other formats are “proprietary,” and typically can be used only by the owner of the format or one of their licensees. Adobe originally developed the PDF format, but released it as an open standard in 2008. This is why so many applications, including open-source applications, can now create, access, and otherwise use PDF files.

As a general rule of thumb for organizing open- or generic-format files, I try to store them where I can logically find them. E.g., research-related files are stored in ~/Documents/Research/*. If multiple application programs can use a particular file type, I either keep this type of file in folder(s) dedicated to a specific purpose and/or a general, application-agnostic folder . For example, I keep PDF files of my tax returns in “~/Documents/Personal/Financial/Taxes,” but I store all scholarly PDF files on my (mounted) NAS drive in myNAS/mnt/Volume1/Documents/ZoteroLibrary.

At first glance the folder name “ZoteroLibrary” might seem to contradict what I said about application-agnostic folders. But I use the Zotfile Zotero plug-in to maintain the library, and the plugin names the files according to a user-specified convention (e.g., “Author Date - Title”) and then simply uses the native file system to organize the PDF files. Any other app can find and open the file by simply following the appropriate path, e.g.: “myNAS/mnt/Volume1/Documents/ZoteroLibrary/AdamsP/AdamsP 1998 - Network topologies and virtual place.pdf”. To access the file one could either know its absolute path or (using Finder, shell commands, or another file navigator) navigate to the folder and access the file.

Zotero & Zotfile organize my PDF files in a logical structure that can readily be used by any app that can read PDF files. This includes Obsidian. But Obsidian’s file organization is designed for note-taking, not library maintenance. Anyone using Obsidian’s vault structure would have to know details about what and how Obsidian does things. IMO, it’s best to avoid such application-specific conventions unless they are what ordinary humans would ordinarily do anyway.

1 Like

Hi, there,
I would like to ask if you still use the “highlights” PDFs reader?

how you manage to extract the inline notes separate from the highlights? or you use the “comments”?

Thank you very much!

This is very useful! Thank you for sharing!

Can I used that on IPAD pro? It realy very helpfull …. Thanks

Did you, or anyone else, try out Protolyst yet?

My workflow

I use Zotero and Obsidian for academic purposes. I have to read many articles, take notes out of them, understand concepts, summarize and synthesize them and then be able to retrieve where they’re coming from.
I like the Zotero notes because they keep the information strictly connected to the article itself, I like the obsidian notes because I can connect different concepts together and reference different sources in a same note.

What I do is the following:

  • Download an article to Zotero
  • Print it and read it
  • Take some notes on a notepad
  • Copy and elaborate the notes on Zotero
  • Export the Zotero note to Markdown with Zotero Better Notes
  • Open the note in Obsidian and fit that information in other related ones

The advantage of this is that the original notes I took from an article are always available both in Zotero and Obsidian and kept in sync. I only create one note per article, but you can probably make more. This is nice because when writing a note on Obsidian I can easily reference what I jotted down from a specific article and when I’m reading them I can easily retrieve what I was thinking about an article. On the Zotero side, this is useful because the original note about an article is always readily available when I open the article again in the library and I don’t need to open a separate application and search through it.

Pain points

  • Images can only be added through Zotero
  • Links get escaped by Zotero and become unusable in Obsidian
  • YAML header fields get reset by Zotero

this is very helpful. Thank you so much. Make my life easier!!!

There’s only plaintext without link after dragging to obsidian, anyone facing the same problem here?

Update: My problem was solved by enable the rich text/HTML: Include Zotero links. For me, just enabling the Markdown: Include Zotero links not works. What you need to change is as follows(Here provides a chinese ver.):

Hope this can help someone facing the same problem.