Today I’m staying with the theme of managing bibliographies (see my previous post on syncing .bib files). Now I describe the process of how I actually add papers to my bibliography, and keep track of everything that I’ve read. This process is fairly recent so this post is rather an exercise in me thinking out all the steps, but perhaps it might be useful to somebody else.
Most of my ideas for which papers I should read come from Google scholar alerts or Twitter. When I think “this might be interesting”, I immediately save the PDF (if I can access it) or the link to my Evernote inbox.
As part of my weekly review, I go through all the notes in my Evernote inbox. When the note is a paper, I decide whether it’s really something I want to read, and if yes, I now definitely get the PDF and put it into the note. I also rename the note by its Bibtex key, for example “cheplygina2017transfer” for a recent paper of mine where the first word in the title is “transfer”. I tag the notes with different keywords, which tell me what topics the paper could be relevant for.
I then move the paper to either “Literature Inbox” or “NextPaper Inbox” notebooks, where NextPaper is the “codename” of the paper I’m going to submit next.
OK, so now the papers are collecting in these two literature inboxes – what next? I need to actually read some papers. Recently I’ve been scheduling tasks like “Process 5 papers” on my calendar to get this done. When I do this I usually select papers which have similar tags, which I would write about in the same section of my paper.
Then I go through the paper and type notes in Evernote, in the same note that already has the PDF. This (the note writing) is inspired by Dr. Raul Pacheco-Vega’s “PDF to memo” method.
This should give me an idea of what I want to say about the paper, if I reference it later. I might also add or remove tags as needed. I can also add Evernote links to other related papers, although I don’t do this very often.
Once this is done, I move the note to “Literature Processed”. This is also when I get the bibliographic details of the paper (by searching the title on Google scholar), and add them to Jabref (by copying and pasting the .bib details). It is also possible to add a link to Jabref that will open the corresponding Evernote note, but unfortunately this uses the URL field, so you wouldn’t be able to have links to both Evernote and the online source of the paper.
I also schedule tasks like “Literature Processed to 0” which means I have to actually include the processed papers in the paper I’m writing. I go to the section where I want to reference the paper and write something about it, based on the notes I made before. This is also a good check to see if I really have the biliographic details – if I don’t, ShareLateX will not suggest me the reference.
Finally, I tag the note with “NextPaper” since that’s where I referenced it, and move it to the “Literature Reference” notebook. I should review these tags upon publication of the paper (I haven’t gotten to that part yet since I started using the system). But if the tag stays, I could even include a short snippet of what it is that I said about the paper.
In a next project I might want to reference some of the same papers that are in Reference, and do not go through the process above. However, once I’m working on “NextPaper2”, I can just go to my Reference notebook, do a search for the relevant tags, and then tag those papers with “NextPaper2”.
This is not a perfect system, since several things to be updated manually. Perhaps Mendeley or Zotero sort out these things for you – I’ve tried both in the past and was never quite satisfied, but this was before I had this process in place.
The process – not the tools – is probably what is helping me the most. But another advantage is that I’m using tools I already use a lot. Since I have Evernote open all the time, it’s easy to decide to read a paper, when I have a bit of time before my next meeting, for example. It also gives me a overview of how much reading/writing I need to do, based on the number of notes in each notebook, and it’s rewarding to see the “Inbox” numbers go down. I think I’ve got the ingredients to successful habit adoption right here – now hoping these will in fact translate to written AND published papers.