Now that I know zotero plugin will still be able to cite with only the metadata (and I’ve tested this), I’m slowly being converted to preferring to import the files into DT and avoid the potential risks of indexing.
What’s stopping me is that I only seem to be able to import the files from a folder as a one-time action on whatever files are there at the moment. Is there a setting or a smart rule that would allow DT to watch the folder into which I place zotero attachments, and automatically import them into DT whenever a new one is added? If so, how do I set this up?
Thanks! I very much appreciate all the advice and suggestions offered here.
Ah, yes, that works just fine! and it has the advantage of giving me pause to decide whether I want to file the items at all. Thank you.
Ok, with the help of everyone here I think I’ve found my workflow, and it does everything I wanted in a simple manner:
Capture items from the web using Zotero’s best in class (for my purposes) browser extensions.
Use the Zotero 7 zotmoov plugin to automatically move all items into DT inbox, thereby avoiding the indexing risks.
Organise the items in DT, using DT’s superior functions for keeping and searching through them.
Still have the metadata in Zotero so that when I write in word I can use the Zotero citation plugin.
So now whenever I have an item to add, there are just two steps and I’m done - one to capture it using the zotero extension, and then one to decide where to file it in DT.
I follow a similar workflow except I use Bookends on the Mac instead of Zotero.
My main difference is I also use Hookmark to link my Bookends reference to the paper that was imported into Devonthink. With a few keystrokes, I have a bi-directional link between the two items, plus anything else I need to hook to for the work I may be doing (Obsidian note, Things task, an outline etc). I added a metadata field in Devonthink to hold the Bookends reference URL and this is working great for me.
I have a mishmash of thoughts and it’s too early in the morning and I’ve not had my second cup of tea yet so I’m just going to dump them as separate points with no logical thread between them. They’re not really related thoughts, beyond relating to what you want to do.
[This point isn’t really for OP, but reflects a remark made in this thread that I want to counter for other readers.] I have strong [negative] feelings about any PDF viewer that handles PDF annotations in a proprietary format, and therefore disagree with anyone who thinks it’s a good idea I’ve written lots about it generally over the years and I won’t re-hash old ground, but as a user I feel the bear minimum we should expect is that our PDF annotations should be in a standard* format that is accessible to all sensible software.
[*There is no agreed standard format yet. But there are accepted mechanics for how annotations are stored in PDF files, and enough apps abide by that to make it the de facto set of principles.]
If you are not doing anything special in your PDFs (I.e. just highlighting, making notes, all normal actions) you should demand -and you deserve- standardised functionality! (There are exceptions, and apps that can justify that. Drawing on PDFs, extensively scribbling - these actions need vector layers and apps like GoodNotes and LiquidText do this well. But for most academic work, standardised PDF annotations are your friend. You’re in this for the long haul. DECADES. What happens if your proprietary app stops work before you do? What if the software changes in a way that ‘breaks’ your workflow and you have to switch to different software? What if there’s a catastrophic failure somewhere?)
I’m not in the academic publication grind, but I am a scientist and I read and cite. I know it’s inefficient, but I do it manually and have done for years now. I probably would feel differently if I was publishing a lot. I like the control and the thinking space that managing citations affords me.
Similar to the above point, papers not downloading with the helpful filenames doesn’t really bother me (and really the problem here is at source - publishers, please stop uploading PDFs with useless names like SE29394ury3f2q8!). For my workflow, processing these PDFs as they hit my inbox is part of my work. I change the filename to my convention (names, year, title), add appropriate tags, decide the file’s priority for my to-do list (assuming I’ve not already read it or it’s not an immediate read), determine its location in my database, maybe create its corresponding markdown file if I already have some thoughts I want to remember when I return to the file. The manual processing of the file affords me the space for this initial bit of thinking.
(Washing up dishes can be a drag, but we never consider what is lost if we switch to a dishwasher and no longer have that quiet 15 mins with hands in hot, soapy water focused on a methodical, manual task. I wash up by hand if you’re wondering )
DT is great for knowledge work and as evidence by this thread there are myriad ways to use it. It plays nicely with the main citation managers andI suspect most of us found our current set-ups mostly through trial and error as we tweaked the bits we like and tried to eliminate the bits we don’t like. This makes it scary for new users, but really you can start small (“this is my problem, here is my solution”) and refine as you go.
I’ve written lots about it generally over the years and I won’t re-hash old ground, but as a user I feel the bear minimum we should expect is that our PDF annotations should be in a standard* format that is accessible to all sensible software.
)