Academic workflows revisited

It’s been a year since I wrote this post on academic workflows. The way I work has changed a bit since then – some apps have gone out, some have been added, and the relationship between some of them have changed, primarily affecting my literature review work flow. Here’s a little post on my current set up. 

The data analysis process 

This is pretty identical to what I wrote in my previous post. I still think Inqscribe is great for transcription, Curio (which I’ve written about here) is great for overall management of the writing process, and iAnnotate is great for those first rounds of analysis of my overall data corpus. In addition to Curio, I often turn to NovaMind, which is a mind mapping app with an extensive feature set. 


The writing process

I also handle the writing process pretty much the same way as I outlined a year ago.

I use mind mapping and outlining tools for brainstorming and organising my ideas (which I’ve written about here and here). I use Scrivener for my writing (which I’ve written about here), and import literature notes, data and pdfs as needed. Scrivener is a fantastic tool for writing academic papers, and I can’t see myself giving up on it anytime soon. When their long awaited iPad app arrives, it will be even more useful.

Towards the end of the writing process, I switch to Word. I use Endnote for citations, inserting unformatted citations in Scrivener and then converting them when the text is in Word. 


The literature review process 

Here’s where most of the changes have happened. I no longer use Sente. I still think it’s a very decent app, but eventually I decided to go with Bookends (which I’ve written more about here). There were two main reasons for this. The first is the ability to do a spotlight search of imported attachments and subsequently create a smart group based on the search results. I find this very useful for review purposes, and it’s a feature that Sente doesn’t have. The second reason is the extremely responsive support – the developers get back to you very quickly, and they stay with you until the problem is solved. 

So, what I typically do with my literature is as follows:

Importing literatureI subscribe to email notifications from all the journals in my academic field. From the email, I open the table of contents in Safari, and from there I import the pdfs I want into Bookends using the print feature. I then use keywords (Bookends’ name for tags) to assign each article to the relevant thematic smart group(s). I use Hazel to automatically add a copy of the downloaded pdf to a dropbox folder. This is because I want to keep a ‘clean’ copy in addition to the pdfs that I annotate. 

Reading and marking up journal articlesI then turn to my iPad. I prefer reading and marking up literature on my iPad, rather than on my Air. I can hold it in the same way as a book, and use a stylus to highlight, underline and write notes. The attachments folder in Bookends is synced to Dropbox, which in turn is synced to GoodReader and iAnnotate (which I’ve written about here) on my iPad. I use GoodReader and iAnnotate to mark up journal articles. When I’m done, I extract the annotations from the pdf file, and send them to myself in an email. My email client Postbox comes with good Evernote integration, which allows me to easily convert the email to a note in Evernote. You can also simply cut and paste the text, or use Skim as described by Joseph Flanagan in the comments below. 

Note taking. I have two main hubs for literature notes: Evernote (which I’ve written about here) and Circus Ponies Notebooks (which I’ve written about here).

In Evernote, I put all ‘single source notes’ – those are notes from a particular journal article, book chapter or book. Each source has a separate note, which includes the formatted reference, the abstract, annotations made on my iPad, and my own reflections about the text and its relevance for my work. These notes are gathered in a separate notebook called ‘review’, and organised according to tags that reflect the smart groups I have created in Bookends. Some of the reasons why I use Evernote as a hub for my single source notes, is that it works cross platform – the sync between my Air, iPad, android phone and Windows machine at the office is instant. I also like the ease with which I can organise and share my notes, and that I can create link between notes. Once the notes have been written up, I return to Bookends and use labels to mark the relevant reference with “in EN”, to make it visible that this reference has a set of notes and annotations in Evernote. 

I use CPN for reviews that are organised according to theme, rather than author/single text. I find it superior to Evernote for this purpose, due to the outlining feature of the notes pages. Outlining makes it easy to create different hierarchical levels for different kinds of information (theme, author, findings, and so on). It’s also convenient for navigating long documents, because you can expand and collapse information as needed. I also create links between different cells and pages, so I can easily refer back to things I’ve already written and connect related content. 

I export notes from both of these apps into Scrivener when I need them during writing. 


…so – that’s more or less my current set up. If you have suggestions for good academic work flows, please leave a comment below.


Related posts on this blog: 

Academic workflows: When to use what, and how

Criteria and questions for establishing your academic workflow

This entry was posted in brainstorming, data analysis, iPad, literature review, mind mapping, outlining, pdf annotation, reference management, transcription, workflows, writing and tagged , , , , , , , . Bookmark the permalink.

6 Responses to Academic workflows revisited

  1. Hi,
    I’ve been rethinking my academic workflow, and have been finding your blog really useful. Perhaps because I grew up on the “one note per index card” method, I find e-mailing myself the annotations to be a pain, because then I manually have to separate them. So right now I’m using either Sente app on the Ipad and then running a script on the Mac that exports the notes to RTF or, in cases where Sente can’t handle a PDF, exporting an annotated PDF (I use PDF Expert), converting the notes to Skim notes, and then running a script that exports the notes. The individual notes get tagged and are indexed in DEVONthink. (I haven’t quite worked out the next process: the sync feature in OmniOutliner works great now, and I’ve only heard negative things about Circus Pony Notebooks on the iPad (I only have the Mac version).

    As I said, I’m still working on this workflow, so I was wondering how well working with all the notes in one file works for you. My feeling is that that it would work fine for the individual article but would be difficult when doing a literature review (I haven’t tried linking the individual notes in Circus Pony, and I’m trying to imagine how I would remember what to link them to). It seems you are using the search function in Bookends for this, so I’d be curious to hear more about how you’re doing it. I’d also be curious how good Bookends is at recognizing different PDF formats. Sente has been a bit flaky for me on this issue.

    • macademise says:

      Thanks for the comment, Joseph, really nice to get some different workflows visible here on the blog.

      I see your ‘one note per index card’ frustration, and I basically agree with you: it works very well with individual articles and not so well with aggregate reviews. I basically do a lot of manual cutting and pasting – which I’m not happy about, but hadn’t found a better way around it until you commented here. Could you please share how you convert to skim notes? Do you use an apple script?

      I generally prefer outlines to tagged index cards/ notes for proper reviews, because I feel it provides a bit more structure and sense of relationship between the different parts. I think OmniOutliner is brilliant (especially now that it finally syncs well), but ended up with CPN for this purpose, for two main reasons: The first is that I find it easier to move back and forth between different overviews in CPN, because they’re all gathered in the same notebook (i.e. the same file). When I used OO, that meant moving between different files. In CPN, I open a file called ‘reviews’, and there are my reviews of “everything”, with one page dedicated to each thematic area/ theoretical field/ question of inquiry and so on. This also means (and here’s the second reason why I prefer CPN) that I can link between different reviews and cells as needed and basically achieve the ‘index cards with more than one tag’ effect, but maintain the structure that comes with outlines.

      But yes, the downside is that the iPad version is really disappointing. I have no idea why they don’t invest a bit more into it, especially with that price tag. I can live with it because I primarily use the iPad version for reference, rather than text production. Writing on it is a pain, reading I can deal with.

      I’m not quite clear on the question with regard to the search function in BE? Basically, I don’t deal with literature notes in BE – only the pdfs themselves.

      Like Sente, I also find BE a bit flaky on importing meta data. While EndNote is otherwise clunky and annoying, I have yet to find something that provides me with the same level of accuracy for importing metadata and doing CWYW. I therefore use BE for pdf management (for that purpose, it doesn’t matter so much if the reference info isn’t 100% correct), and EndNote for CWYW. It’s not a perfect set up – but then, there isn’t a perfect app for this, either. I’m planning to give writing a full paper with BE a try soon, though.

      • About the BE question first: it was in reference to how you are using spotlight search for imported attachments in BE and then creating smart groups. It was mentioned under the literature review process, so I thought you were using it to write aggregate reviews.

        I haven’t found spotlight searchers to be much of a problem with Sente, as just about anything I would probably search for in a PDF winds up in a note in a RTF format that both Spotlight and DEVONthink can find. (I name the title with something informative about the specific note, put it in a folder on a related theme, and then use a Hazel rule to append additional information so I can find it more easily (so, for instance, original name might be “meta-analysis – morphosyntax” and then the final name might be something like, “refx – interaction – meta-analysis – morphosyntax” – where the “refx – interaction” was added by the Hazel rule. This is probably overkill (especially with the tags), and might be hard to sustain, but, as I said, I’m just starting thinking about what works and what doesn’t when I want to find it later.

        The script is indeed an Apple script: Rob Trew’s script for Skim to DEVONthink, located here

        (Sorry, I’m not sure of the best way of handling URLs in comments in WordPress, so I hope that turns out OK):

        I just picked up Keyboard Maestro, and I’m wondering whether it would be possible to create a macro that mimics the ability to export the notes individually (I don’t know either app well enough at the moment to know if it is possible, but Keyboard Maestro might really open up possibilities once I get a handle on how it works).

        I’m still on the fence if I want to create a themed notebook in CPN. I’m really in the stage right now where I’m trying a lot of different methods and seeing which one works the best. I’ll probably give it a go, but. of course, the biggest problem is that all this tinkering with workflow prevents one from actually doing work :-).

        • macademise says:

          Right, I see. Yes, I use the search function in BE for creating aggregate reviews, but it’s usually as an extra check on which pdfs to review. While I have a system of smart groups and tags that work well for me, I might do a spotlight search on a specific concept which I don’t necessarily have a group/tag for, and which therefore might exist in more than one group or across a few tags. Or, I might do a search for every pdf that have cited a particular author. That kind of thing. While I could do this in Devonthink or Finder, I like being able to create a smart group based on the search results in BE, so that I have the meta data easily available when I review the results.

          Many thanks for the script! I’ll definitely look into that. I’d quite like a solution to the breaking down of notes from individual sources. More generally, there are four things that I haven’t had the time to look into properly, which I think represent a lot of untapped potential: one is the use of apple scripts (which I still feel very incompetent dealing with), and the three others are Hazel, Automator and Keyboard Maestro. I have all these apps, but only use them for really basic stuff at the moment, and I’m sure I could do a lot more with them. But my work load will be quite heavy until the spring, so I’ve had to put some of that experimentation and tinkering aside for now.

          I think what you do with RTF files sounds very interesting, and I’ve heard of a lot of similar approaches to organising review notes, especially in conjunction with Devonthink. There’s really not a fixed answer to these workflow questions, so trying out different things is the only way to go. I don’t know how many hours I’ve spent on the “tinkering”… :/ … better not to think about that! 🙂

          Thanks for sharing your workflow here on the blog!

      • Just to clarify, the script is for exporting Skim Notes to DEVONthink. You don’t need an AppleScript to convert the highlights on a PDF to Skim Notes. Just select Convert Notes from the File menu in Skim. From there, you can either use the Skim to Devonthink script, or export all the notes as a single RTF, or copy an individual note and paste it where you wish. That’s where I think it should be possible to use Keyboard Maestro (that is, some method of copying a particular note, pasting it into another application, and then moving to the next note and repeating the same process: like you, however, I probably won’t have time to do this for a while). Even without using a macro or a script, it might be faster for you to work with Skim notes than working with an e-mail that contains all the annotations together.

        • macademise says:

          Sorry, that was a typo on my part. I just tried the convert notes in Skim. I’ve used that before, ages ago, and had forgotten about it – thanks for the reminder.

          What I used to do for a while for individual source notes (which I left out of this post, but I’ve integrated it in a future post on Devonthink), was to use the Bookends template in DT to create a rich text file linked to the BE reference. I did a simple cut and paste of the annotations I emailed to myself, and added my own notes – the reference information, keywords and abstract got imported into the RTF in DT automatically. When I was done with the note, I then used the DT export function to send a copy of the text to Evernote (which would also transfer the tags). This was for single source notes (not aggregate reviews). I also synced the RTF files in DT to a dropbox folder, to make them available on my iPad. If I didn’t use an iPad, I would probably have kept this set up. However, I didn’t like working with the RTF files on my iPad, because it would often mess up the format of the template, and I couldn’t see the tag structure on my iPad (I guess this is why a lot of people prefer plain text and markdown – but I’m just too fond of my colour codes and so on). While the notes were copied to Evernote (which made them very easy to access and edit on my iPad), I didn’t like to have two version of the notes, as I sometimes update notes as I work with them. Hence, DT was taken out of the equation.

Comments are closed.