I am now starting up my PhD in linguistics. I have already collected more 1500 PDF articles and books (also did my MA in linguistics). So, I am trying to build up as perfect workflow as possible for my future research works. The university has given me a macbook pro, so , I am no more using windows OS. Even if there doesn´t seem to exist any comparable application in Mac as MS OneNote, I am discovering quite powerful apps in mac OS too. I have already learned a lot about Devonthink, Cirus Punies Notebook, Curio, Tinderbox and the like great apps. Therefore, I will be recording my experiences with each of the apps I am trying until I come up with the final, perfect system for my work flow. I will write a detailed review of each of the applications here in the future. But, for now , I will just put only a short summary of my experiences with them.
1. File Organizer
My first task is to properly setup files organized in a specific folder, to make them easily accessible via Spotlight (or Alfred, I prefer the latter though). For file, organization, I use two tools; Dropbox folder and Mendeley. Dropbox doesn´t require explanation. I use mendeley not only to collect references, but also rename and organize my PDFs. It is such a powerful application to do these tasks. Here I use it.
a-I set a folder in Finder, I call it “agglomeration”, to mean, a folder where I drop all newly downloaded PDFs. All the PDF I download from internet directly go there. I use a download manager called folx to force all the pdf files to go to this folder. You can google it.
b-I have another folder in Dropbox, call it “AllLing”. This is the folder where I keep properly organized files.
c. Then, I setup Mendeley to suck-in all the PDFs available in the “agglomeration” folder into its library, rename and then put them all into “AllLing” folder.
As you can see from the above screenshot, the Mendeley is organizing my PDFs into a folder, inside Dropbox. Since the files will be renamed to Author-year-title, I can search the files using any of these attributes. I also index the folder “AllLing” into Devonthink (see next). One main reason I want to use Mendeley is the fact that it live syns Bibtex files , even if it is not as elegant as I wish it to be. Other reference managers such as Sente and Papers are great by their own, but are weaker in their integration with bibtex.
Database managers are tools to organize files and information in a manageable manner. I use Devonthink this purpose. Devonthink is one of the most powerful apps I have ever seen in the mac environment. It has an artificial intelligence which looks inside the PDF files and establish content-based relation among the PDFs. That means, if I am reading an article on “Definiteness” , the software can scan its database, find and suggest relevant articles, articles that contain the word “definiteness” or/and other related words in the articles for me. It is also packed with many other interesting features such as tagging system, notes-taking tool; organize files into different folders, smart folders, duplicate detection, replication (aliases) etc. If you are staring to use the app, the learning could be a bit steep. I definitely recommend you to watch a screen cast in a website (it is under a paywall unfortunately) called screencastonline. Their screencast gave me a good ground on Devonthink. (Note: I don´t have any affiliation with any of the links I mention here). Devonthink will be an indispensable part of to my workflow. I have tried some of the other database apps. I think no other app as good as Devonthink for managing scientific papers. Therefore, my database agenda is closed. The challenge I am facing is to make other applications to work with Devonthink.
So, Mendeley renames and puts the files in “AllLing”; Devonthink indexes them. I then group, replicate, organize, tag the files in the Devonthink so that I could organize them for my specific projects. I am right now writing a paper about Nominalization. Hence, I search and “see also” the related papers in Devonthink, Group them in one folder; I then drag them to Sente for reading and taking notes.
3. PDF annotation and note-taking:
Macadamec has already written a great post about Sent. I recommend visiting his post; I am not going to repeat the whole story here. I will just shortly reflect my own experience with the application and its place in my work flow.
I am considering totally leaving Mendeley and migrating to Sente because of the fact that the application has a more elegant tools of annotating PDFs. It can directly quite, snapshot, highlight and insert all these into the Notes panel. That is brilliant. It can also rename files, just like Mendeley. The notes then could be exported to Devonthink or Scrivener using some apple scripts. brilliant!
Sente has some fundamental flaws, unfortunately, that makes me nervous to totally migrate my data from Mendeley:
a. it fails to import PDFs from other applications,
b. the link between the note and the pdf could also be broken. Some people have experienced this problem, and I had the same issue with a few PDF annotations. Right now, I am using it only as PDF annotation, not as a reference manager. My references and PDFs remain in Mendeley while I temporarily import the PDFs I want to read into Sente. (just search in Alfred and drag it to Sente because the files are properly renamed by Mendeley, or go to the folder “Allling” and drag the file; but, I usually drag them from Devonthink).
c. it is also bad for Latex integration
d. Annotations are not stored in the PDF: this the problem of almost all the note-taking tools in Mac; they store the annotations in their own database. If you open a PDF from Dropbox in another PDF reader or browser, you couldn´t see the annotation done in Sente (or Mendeley or Papers) while the annotation done in Acrobat or Foxit or PDFexchange are there, everywhere you have the pdf. Storing the annotation is good for long term use, as these applications could break. But, Sente couldn´t do it, unfortunately.
Here, the choice is clear. Since all the notes are exported from Sente in either OPML or RTFD format, I just import them to Scrivener.
5. Final Polishing and Publishing:
I export my draft form Scrivener in Latex format, I import the text to TexStudio, a latex editor that I use to finally polish my work. Texstudio, and also TexShop, can automatically detect and insert my references which are stored in Jabref (in sync with Mendeley).
Finally, a shinny PDF!