Groups/Folder structure for Literature Review


I’m a doctoral student working on my dissertation literature review. I’ve recently started using Devonthink Pro to organize materials and would like to use it to capture all my notes and quotations as I review the literature in my area. I’m looking for advice on setting up a good, sensible group or folder structure for each article I review. Does anyone have any experience they would like to share? I’m thinking about breaking down the review by outline (creating a folder for each topic) and keeping my pdfs and notes in these folders arranged by topic or section of the literature review. Any thoughts will be greatly appreciated!


I’m an academic as well and I keep all of my journal articles (over 1GB) indexed in DT 2.0, but they are sorted by journal title. I suppose the reason for this is that DT’s search capability pretty much negated the need to sort them based on topic, plus I couldn’t get straight in my own head what to do if, ostensibly, one article clearly belonged in two or even three topic/themes folders. Anyway, sorting by journal title has worked wonders for me, and I’ve even been able to find, on occasion, a random thought or piece of analysis that I remember reading but, for the life of me, just couldn’t remember where. Using search, was able to find within 5 min based on combinations of search phrases.

What you could do is have a centralised master folder with the articles you are reviewing, then replicate each to a separate folder elsewhere that not only contains the replicant, but the .txt or .rtf file of your review/quotes/etc.

Finally, unless you have the need for complex mathematic equations (I sometimes do, but not all that often), try doing the bulk of your writing within DT. I’ve actually found it to be excellent, and have written numerous chapters and articles directly in it before copying out to Word for any final formatting that might be required.

I use a system very similar to what Chatoyer has described. I found it works wonders, and used the system to write my literature reviews. I, too, have found that using the journal article name works well and while I had hesitated earlier about storing PDFs in DTP, I’ve now broken down, done it, and will attempt this afternoon to use DTPs external linking system to copy links to the files into Bookends instead of storing articles in multiple places.

This has been the bane of my existence (OK, I exaggerate, but only a little :wink: ) since I’ve started using DEVONthink. Of course, as you identify, DT’s excellent searching means that one big folder would be only marginally less efficient. However, what I think I’m going to do is what you have suggested: organise by journal. Then, when DEVONtech implement tagging, I’ll add the topics as tags, and use smart folders to browse by topic. If I need to regularly access a specific set of journal articles for an ongoing topic, I’ll replicate/duplicate them to a new, project specific folder.

One suggestion that I would make before developing an elaborate group/folder structure today would be to wait for tagging to be fully implemented in DTP 2. Tagging could, in all likelihood will, make a big difference in how users develop their database structure. Also, FWIW, unless the DEVON folks have changed their thinking, tags in DTP 2 will be replicants.

I’m not 100% sure what you mean by that. Could you elaborate?

Tagging files in DTP will be essentially the same thing as creating replicants and folders will represent the tag groups. The basic structure for this is already in place in DT 2. To demonstrate, look at the info panel for any items in your database that have replicants. You’ll see that the tags assigned to the replicants match all the folders (tags groups) where the replicants are located in the database. Folders, and all files located within that folder, can be excluded from tagging by selecting the folder and checking ‘Exclude from… Tagging’.